Token | announced |
Token ID | 3414 |
Feature activation | +4.85e+00 |
Pos logit contributions | |
arthed | +1.465 |
ept | +1.370 |
##### | +1.345 |
iltr | +1.313 |
Dol | +1.302 |
Neg logit contributions | |
ibaba | -1.165 |
racuse | -1.110 |
Vector | -1.107 |
WARE | -1.083 |
buquerque | -1.070 |
Token | through |
Token ID | 832 |
Feature activation | +1.23e+01 |
Pos logit contributions | |
arthed | +3.212 |
ept | +2.950 |
##### | +2.899 |
goodbye | +2.884 |
Dol | +2.837 |
Neg logit contributions | |
ibaba | -3.281 |
racuse | -3.169 |
Vector | -3.119 |
WARE | -3.118 |
buquerque | -3.056 |
Token | the |
Token ID | 262 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
arthed | +7.204 |
Dol | +6.711 |
ept | +6.688 |
##### | +6.614 |
goodbye | +6.430 |
Neg logit contributions | |
ibaba | -8.499 |
racuse | -7.976 |
WARE | -7.869 |
buquerque | -7.722 |
sidx | -7.624 |
Token | state |
Token ID | 1181 |
Feature activation | +1.22e+01 |
Pos logit contributions | |
arthed | +6.551 |
Dol | +6.387 |
##### | +6.372 |
ept | +6.328 |
goodbye | +6.124 |
Neg logit contributions | |
ibaba | -6.848 |
racuse | -6.601 |
buquerque | -6.248 |
WARE | -6.198 |
Vector | -6.081 |
Token | news |
Token ID | 1705 |
Feature activation | +1.28e+01 |
Pos logit contributions | |
arthed | +6.999 |
ept | +6.788 |
Dol | +6.516 |
##### | +6.503 |
goodbye | +6.407 |
Neg logit contributions | |
ibaba | -8.155 |
racuse | -8.071 |
buquerque | -7.861 |
Notting | -7.728 |
Vector | -7.673 |
Token | agency |
Token ID | 4086 |
Feature activation | +2.32e+01 |
Pos logit contributions | |
arthed | +5.603 |
ept | +5.275 |
##### | +4.817 |
Dol | +4.704 |
goodbye | +4.434 |
Neg logit contributions | |
ibaba | -10.614 |
racuse | -10.458 |
sidx | -10.374 |
Notting | -10.096 |
Vector | -10.056 |
Token | KC |
Token ID | 25247 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
Dol | +12.026 |
ept | +11.617 |
##### | +11.444 |
arthed | +11.193 |
goodbye | +10.514 |
Neg logit contributions | |
ibaba | -14.390 |
racuse | -14.294 |
buquerque | -13.499 |
WARE | -13.134 |
sidx | -12.586 |
Token | NA |
Token ID | 4535 |
Feature activation | +8.71e+00 |
Pos logit contributions | |
ibaba | +1.049 |
Vector | +1.021 |
racuse | +0.997 |
WARE | +0.994 |
buquerque | +0.974 |
Neg logit contributions | |
arthed | -1.129 |
ept | -1.051 |
##### | -1.027 |
iltr | -1.006 |
Dol | -0.977 |
Token | , |
Token ID | 11 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
arthed | +5.197 |
ept | +4.857 |
goodbye | +4.749 |
##### | +4.696 |
Dol | +4.638 |
Neg logit contributions | |
ibaba | -6.075 |
racuse | -5.932 |
Vector | -5.751 |
Notting | -5.691 |
WARE | -5.683 |
Token | that |
Token ID | 326 |
Feature activation | +1.25e+01 |
Pos logit contributions | |
arthed | +6.044 |
ept | +5.636 |
goodbye | +5.601 |
##### | +5.549 |
Dol | +5.484 |
Neg logit contributions | |
ibaba | -6.936 |
racuse | -6.879 |
WARE | -6.675 |
Vector | -6.582 |
buquerque | -6.524 |
Token | tourism |
Token ID | 19277 |
Feature activation | +7.17e+00 |
Pos logit contributions | |
arthed | +6.890 |
goodbye | +6.835 |
Dol | +6.693 |
##### | +6.640 |
ept | +6.510 |
Neg logit contributions | |
ibaba | -8.636 |
racuse | -8.583 |
buquerque | -8.195 |
WARE | -8.192 |
Vector | -7.867 |
Token | a |
Token ID | 257 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
arthed | +2.588 |
ept | +2.424 |
##### | +2.401 |
Dol | +2.310 |
goodbye | +2.288 |
Neg logit contributions | |
ibaba | -2.295 |
racuse | -2.188 |
Vector | -2.162 |
WARE | -2.133 |
buquerque | -2.111 |
Token | heap |
Token ID | 24575 |
Feature activation | -6.74e+00 |
Pos logit contributions | |
arthed | +1.336 |
ept | +1.259 |
##### | +1.240 |
iltr | +1.197 |
Dol | +1.191 |
Neg logit contributions | |
ibaba | -1.147 |
Vector | -1.092 |
racuse | -1.090 |
WARE | -1.066 |
buquerque | -1.056 |
Token | like |
Token ID | 588 |
Feature activation | +5.40e+00 |
Pos logit contributions | |
Vector | +3.565 |
ibaba | +3.562 |
racuse | +3.315 |
WARE | +3.308 |
Higgins | +3.305 |
Neg logit contributions | |
arthed | -5.346 |
ept | -4.916 |
iltr | -4.817 |
##### | -4.746 |
Dol | -4.573 |
Token | what |
Token ID | 644 |
Feature activation | +6.41e+00 |
Pos logit contributions | |
arthed | +3.666 |
ept | +3.439 |
##### | +3.382 |
Dol | +3.297 |
goodbye | +3.262 |
Neg logit contributions | |
ibaba | -3.489 |
racuse | -3.382 |
Vector | -3.310 |
WARE | -3.265 |
buquerque | -3.261 |
Token | happened |
Token ID | 3022 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
arthed | +4.020 |
ept | +3.769 |
##### | +3.727 |
Dol | +3.639 |
goodbye | +3.533 |
Neg logit contributions | |
ibaba | -4.401 |
racuse | -4.323 |
Vector | -4.166 |
WARE | -4.165 |
buquerque | -4.162 |
Token | at |
Token ID | 379 |
Feature activation | +2.17e+01 |
Pos logit contributions | |
arthed | +5.956 |
##### | +5.622 |
ept | +5.576 |
Dol | +5.531 |
goodbye | +5.481 |
Neg logit contributions | |
ibaba | -7.467 |
racuse | -7.211 |
WARE | -7.093 |
Vector | -6.904 |
buquerque | -6.857 |
Token | Three |
Token ID | 7683 |
Feature activation | +3.14e+00 |
Pos logit contributions | |
Dol | +11.165 |
arthed | +11.110 |
##### | +10.804 |
Chern | +10.355 |
ept | +10.353 |
Neg logit contributions | |
ibaba | -12.854 |
WARE | -12.474 |
racuse | -12.403 |
buquerque | -12.155 |
ABLE | -12.002 |
Token | Mile |
Token ID | 26004 |
Feature activation | +6.82e+00 |
Pos logit contributions | |
arthed | +1.898 |
ept | +1.799 |
##### | +1.760 |
Dol | +1.689 |
iltr | +1.679 |
Neg logit contributions | |
ibaba | -2.309 |
racuse | -2.223 |
WARE | -2.199 |
Vector | -2.187 |
ABLE | -2.144 |
Token | Island |
Token ID | 5451 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
arthed | +4.417 |
##### | +4.083 |
ept | +4.036 |
Dol | +3.886 |
goodbye | +3.862 |
Neg logit contributions | |
ibaba | -4.686 |
WARE | -4.440 |
racuse | -4.424 |
Vector | -4.346 |
ABLE | -4.276 |
Token | . |
Token ID | 13 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
arthed | +6.138 |
##### | +5.830 |
ept | +5.752 |
goodbye | +5.565 |
Dol | +5.534 |
Neg logit contributions | |
ibaba | -7.092 |
racuse | -6.859 |
WARE | -6.675 |
ABLE | -6.504 |
Vector | -6.459 |
Token | � |
Token ID | 447 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
arthed | +0.162 |
ept | +0.152 |
##### | +0.149 |
iltr | +0.146 |
Dol | +0.143 |
Neg logit contributions | |
ibaba | -0.125 |
Vector | -0.119 |
racuse | -0.118 |
WARE | -0.115 |
buquerque | -0.114 |
Token | ( |
Token ID | 357 |
Feature activation | +4.92e+00 |
Pos logit contributions | |
arthed | +0.549 |
ept | +0.517 |
##### | +0.507 |
iltr | +0.495 |
Dol | +0.486 |
Neg logit contributions | |
ibaba | -0.444 |
Vector | -0.424 |
racuse | -0.421 |
WARE | -0.412 |
buquerque | -0.407 |
Token | L |
Token ID | 43 |
Feature activation | +9.34e+00 |
Pos logit contributions | |
arthed | +3.040 |
ept | +2.799 |
##### | +2.707 |
iltr | +2.624 |
bell | +2.606 |
Neg logit contributions | |
ibaba | -3.577 |
racuse | -3.489 |
Vector | -3.420 |
buquerque | -3.308 |
WARE | -3.274 |
Token | ) |
Token ID | 8 |
Feature activation | +9.37e+00 |
Pos logit contributions | |
arthed | +6.373 |
ept | +5.974 |
##### | +5.935 |
Dol | +5.822 |
iltr | +5.779 |
Neg logit contributions | |
racuse | -5.652 |
ibaba | -5.584 |
Vector | -5.392 |
WARE | -5.288 |
buquerque | -5.125 |
Token | and |
Token ID | 290 |
Feature activation | +8.37e+00 |
Pos logit contributions | |
arthed | +6.083 |
ept | +5.824 |
##### | +5.782 |
goodbye | +5.673 |
Dol | +5.661 |
Neg logit contributions | |
ibaba | -6.127 |
racuse | -6.061 |
WARE | -5.715 |
Vector | -5.553 |
buquerque | -5.529 |
Token | his |
Token ID | 465 |
Feature activation | +5.65e+00 |
Pos logit contributions | |
arthed | +5.397 |
##### | +5.217 |
Dol | +5.115 |
ept | +5.030 |
goodbye | +4.919 |
Neg logit contributions | |
ibaba | -5.618 |
racuse | -5.350 |
WARE | -5.183 |
buquerque | -4.995 |
Vector | -4.966 |
Token | partner |
Token ID | 5212 |
Feature activation | +2.10e+01 |
Pos logit contributions | |
arthed | +3.263 |
##### | +3.150 |
ept | +3.076 |
Dol | +3.010 |
goodbye | +2.992 |
Neg logit contributions | |
ibaba | -4.308 |
racuse | -4.077 |
WARE | -4.013 |
buquerque | -3.939 |
Vector | -3.919 |
Token | Larry |
Token ID | 13633 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
##### | +11.244 |
Dol | +11.101 |
goodbye | +10.761 |
arthed | +10.596 |
ept | +10.194 |
Neg logit contributions | |
racuse | -13.013 |
ibaba | -13.008 |
WARE | -11.778 |
buquerque | -11.287 |
heit | -10.980 |
Token | Mc |
Token ID | 1982 |
Feature activation | +5.23e+00 |
Pos logit contributions | |
arthed | +2.666 |
ept | +2.548 |
##### | +2.530 |
iltr | +2.476 |
Dol | +2.459 |
Neg logit contributions | |
ibaba | -1.012 |
racuse | -0.937 |
Vector | -0.928 |
WARE | -0.896 |
buquerque | -0.861 |
Token | Queen |
Token ID | 32466 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
arthed | +3.817 |
ept | +3.629 |
##### | +3.556 |
iltr | +3.433 |
bell | +3.407 |
Neg logit contributions | |
ibaba | -3.206 |
Vector | -3.067 |
racuse | -3.017 |
WARE | -2.949 |
buquerque | -2.872 |
Token | . |
Token ID | 13 |
Feature activation | +4.88e+00 |
Pos logit contributions | |
arthed | +6.418 |
##### | +6.403 |
goodbye | +6.351 |
ept | +6.267 |
Dol | +6.005 |
Neg logit contributions | |
racuse | -6.739 |
ibaba | -6.666 |
WARE | -6.319 |
buquerque | -6.147 |
Vector | -6.128 |
Token | ( |
Token ID | 357 |
Feature activation | +7.27e+00 |
Pos logit contributions | |
arthed | +3.528 |
##### | +3.348 |
ept | +3.292 |
Dol | +3.224 |
iltr | +3.182 |
Neg logit contributions | |
ibaba | -2.998 |
racuse | -2.909 |
Vector | -2.785 |
WARE | -2.781 |
buquerque | -2.723 |
Token | franchise |
Token ID | 8663 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
arthed | +5.257 |
ept | +5.082 |
goodbye | +5.007 |
##### | +4.920 |
Dol | +4.892 |
Neg logit contributions | |
ibaba | -6.241 |
WARE | -5.948 |
racuse | -5.893 |
buquerque | -5.758 |
Vector | -5.716 |
Token | , |
Token ID | 11 |
Feature activation | +5.10e+00 |
Pos logit contributions | |
arthed | +6.509 |
ept | +6.321 |
goodbye | +6.218 |
##### | +5.817 |
iltr | +5.783 |
Neg logit contributions | |
ibaba | -7.619 |
racuse | -7.265 |
WARE | -7.259 |
Vector | -7.089 |
buquerque | -6.976 |
Token | it |
Token ID | 340 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
arthed | +3.278 |
ept | +3.108 |
##### | +3.014 |
Dol | +2.966 |
goodbye | +2.957 |
Neg logit contributions | |
ibaba | -3.462 |
racuse | -3.329 |
WARE | -3.293 |
Vector | -3.290 |
buquerque | -3.239 |
Token | 's |
Token ID | 338 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
arthed | +2.456 |
ept | +2.351 |
##### | +2.307 |
iltr | +2.239 |
Dol | +2.225 |
Neg logit contributions | |
ibaba | -2.075 |
racuse | -1.985 |
Vector | -1.983 |
WARE | -1.981 |
buquerque | -1.936 |
Token | a |
Token ID | 257 |
Feature activation | +3.84e+00 |
Pos logit contributions | |
arthed | +2.180 |
ept | +2.059 |
##### | +2.015 |
goodbye | +1.973 |
Dol | +1.972 |
Neg logit contributions | |
ibaba | -1.957 |
racuse | -1.894 |
Vector | -1.858 |
WARE | -1.839 |
buquerque | -1.831 |
Token | multif |
Token ID | 43543 |
Feature activation | +2.09e+01 |
Pos logit contributions | |
arthed | +2.552 |
ept | +2.425 |
##### | +2.378 |
goodbye | +2.341 |
Dol | +2.331 |
Neg logit contributions | |
ibaba | -2.542 |
racuse | -2.470 |
buquerque | -2.397 |
WARE | -2.397 |
Vector | -2.387 |
Token | acet |
Token ID | 23253 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
iltr | +9.598 |
ept | +9.381 |
bell | +8.894 |
auna | +8.800 |
##### | +8.612 |
Neg logit contributions | |
WARE | -15.078 |
racuse | -14.694 |
ibaba | -13.867 |
buquerque | -13.838 |
Higgins | -13.803 |
Token | ed |
Token ID | 276 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
arthed | +1.068 |
ept | +1.012 |
##### | +0.991 |
iltr | +0.974 |
goodbye | +0.952 |
Neg logit contributions | |
ibaba | -0.673 |
Vector | -0.637 |
racuse | -0.634 |
WARE | -0.618 |
buquerque | -0.615 |
Token | process |
Token ID | 1429 |
Feature activation | +7.15e-02 |
Pos logit contributions | |
arthed | +1.780 |
ept | +1.688 |
##### | +1.665 |
iltr | +1.618 |
goodbye | +1.616 |
Neg logit contributions | |
ibaba | -1.585 |
racuse | -1.525 |
WARE | -1.493 |
Vector | -1.491 |
buquerque | -1.470 |
Token | . |
Token ID | 13 |
Feature activation | +5.41e+00 |
Pos logit contributions | |
arthed | +0.052 |
ept | +0.049 |
##### | +0.048 |
iltr | +0.047 |
Dol | +0.046 |
Neg logit contributions | |
ibaba | -0.044 |
Vector | -0.043 |
racuse | -0.042 |
WARE | -0.041 |
buquerque | -0.041 |
Token | Organizations |
Token ID | 41846 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
arthed | +3.610 |
##### | +3.475 |
ept | +3.416 |
Dol | +3.399 |
iltr | +3.230 |
Neg logit contributions | |
ibaba | -3.612 |
racuse | -3.426 |
WARE | -3.303 |
buquerque | -3.261 |
Vector | -3.205 |
Token | decide |
Token ID | 5409 |
Feature activation | +5.44e+00 |
Pos logit contributions | |
arthed | +2.595 |
##### | +2.481 |
ept | +2.480 |
iltr | +2.413 |
goodbye | +2.388 |
Neg logit contributions | |
ibaba | -2.469 |
racuse | -2.373 |
Vector | -2.325 |
WARE | -2.304 |
buquerque | -2.285 |
Token | to |
Token ID | 284 |
Feature activation | +6.52e+00 |
Pos logit contributions | |
arthed | +3.398 |
ept | +3.205 |
goodbye | +3.132 |
##### | +3.129 |
Dol | +3.068 |
Neg logit contributions | |
ibaba | -3.873 |
racuse | -3.735 |
Vector | -3.619 |
WARE | -3.568 |
buquerque | -3.557 |
Token | keep |
Token ID | 1394 |
Feature activation | +1.62e+01 |
Pos logit contributions | |
arthed | +4.101 |
##### | +3.995 |
ept | +3.937 |
goodbye | +3.874 |
iltr | +3.839 |
Neg logit contributions | |
ibaba | -4.412 |
racuse | -4.297 |
WARE | -4.171 |
buquerque | -4.112 |
Vector | -4.099 |
Token | an |
Token ID | 281 |
Feature activation | +9.69e+00 |
Pos logit contributions | |
arthed | +8.924 |
Dol | +8.906 |
goodbye | +8.731 |
ept | +8.564 |
##### | +8.534 |
Neg logit contributions | |
ibaba | -10.795 |
racuse | -10.359 |
buquerque | -9.894 |
WARE | -9.891 |
heit | -9.394 |
Token | American |
Token ID | 1605 |
Feature activation | +1.42e+01 |
Pos logit contributions | |
arthed | +6.532 |
##### | +6.160 |
ept | +6.128 |
goodbye | +5.996 |
iltr | +5.993 |
Neg logit contributions | |
racuse | -5.992 |
ibaba | -5.846 |
WARE | -5.816 |
buquerque | -5.638 |
Vector | -5.611 |
Token | tourist |
Token ID | 18473 |
Feature activation | +2.08e+01 |
Pos logit contributions | |
arthed | +7.763 |
goodbye | +7.666 |
ept | +7.339 |
Dol | +7.288 |
##### | +7.264 |
Neg logit contributions | |
racuse | -9.524 |
ibaba | -9.451 |
WARE | -9.256 |
buquerque | -9.029 |
sidx | -8.787 |
Token | on |
Token ID | 319 |
Feature activation | +1.30e+01 |
Pos logit contributions | |
goodbye | +10.208 |
Dol | +9.805 |
##### | +9.549 |
arthed | +9.533 |
ept | +9.283 |
Neg logit contributions | |
ibaba | -14.214 |
racuse | -13.914 |
WARE | -13.369 |
sidx | -12.696 |
buquerque | -12.558 |
Token | their |
Token ID | 511 |
Feature activation | +9.89e+00 |
Pos logit contributions | |
arthed | +7.564 |
goodbye | +7.030 |
Dol | +6.965 |
##### | +6.852 |
ept | +6.852 |
Neg logit contributions | |
racuse | -8.616 |
ibaba | -8.427 |
buquerque | -8.154 |
WARE | -7.987 |
Vector | -7.961 |
Token | territory |
Token ID | 7674 |
Feature activation | +9.35e+00 |
Pos logit contributions | |
arthed | +6.163 |
##### | +6.039 |
goodbye | +6.039 |
Dol | +5.939 |
ept | +5.876 |
Neg logit contributions | |
ibaba | -6.552 |
racuse | -6.392 |
buquerque | -6.049 |
WARE | -6.031 |
sidx | -5.842 |
Token | for |
Token ID | 329 |
Feature activation | +8.29e+00 |
Pos logit contributions | |
arthed | +5.447 |
goodbye | +5.342 |
ept | +5.266 |
##### | +5.233 |
Dol | +5.077 |
Neg logit contributions | |
ibaba | -6.593 |
racuse | -6.409 |
WARE | -6.113 |
buquerque | -6.087 |
Vector | -6.071 |
Token | symbolic |
Token ID | 18975 |
Feature activation | +9.49e+00 |
Pos logit contributions | |
arthed | +5.176 |
goodbye | +4.920 |
##### | +4.876 |
ept | +4.869 |
Dol | +4.837 |
Neg logit contributions | |
racuse | -5.612 |
ibaba | -5.545 |
WARE | -5.302 |
Vector | -5.217 |
buquerque | -5.213 |
Token | meta |
Token ID | 13634 |
Feature activation | +3.21e+00 |
Pos logit contributions | |
arthed | +4.614 |
##### | +4.520 |
ept | +4.377 |
Dol | +4.314 |
iltr | +4.248 |
Neg logit contributions | |
ibaba | -4.323 |
racuse | -4.175 |
buquerque | -4.086 |
WARE | -4.042 |
Vector | -3.995 |
Token | - |
Token ID | 12 |
Feature activation | +5.64e+00 |
Pos logit contributions | |
arthed | +2.284 |
ept | +2.125 |
##### | +2.100 |
iltr | +2.053 |
Dol | +1.997 |
Neg logit contributions | |
ibaba | -2.027 |
Vector | -1.942 |
racuse | -1.937 |
buquerque | -1.910 |
WARE | -1.890 |
Token | reg |
Token ID | 2301 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
arthed | +3.166 |
ept | +2.986 |
##### | +2.905 |
iltr | +2.873 |
bell | +2.723 |
Neg logit contributions | |
ibaba | -4.262 |
racuse | -4.144 |
Vector | -4.112 |
WARE | -4.058 |
buquerque | -4.010 |
Token | ression |
Token ID | 2234 |
Feature activation | +5.09e+00 |
Pos logit contributions | |
arthed | +1.067 |
ept | +0.997 |
##### | +0.976 |
iltr | +0.956 |
Dol | +0.923 |
Neg logit contributions | |
ibaba | -1.150 |
Vector | -1.122 |
racuse | -1.111 |
WARE | -1.098 |
buquerque | -1.086 |
Token | if |
Token ID | 611 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
arthed | +3.465 |
ept | +3.261 |
##### | +3.234 |
Dol | +3.129 |
iltr | +3.108 |
Neg logit contributions | |
ibaba | -3.273 |
racuse | -3.167 |
buquerque | -3.081 |
Vector | -3.021 |
WARE | -3.020 |
Token | more |
Token ID | 517 |
Feature activation | +2.03e+01 |
Pos logit contributions | |
##### | +6.446 |
arthed | +6.398 |
ept | +6.172 |
Dol | +6.094 |
goodbye | +6.038 |
Neg logit contributions | |
racuse | -7.273 |
ibaba | -7.126 |
WARE | -6.988 |
buquerque | -6.804 |
Vector | -6.722 |
Token | than |
Token ID | 621 |
Feature activation | +7.11e+00 |
Pos logit contributions | |
##### | +9.267 |
arthed | +9.061 |
ept | +8.677 |
Dol | +8.462 |
/# | +8.451 |
Neg logit contributions | |
racuse | -13.765 |
ibaba | -12.850 |
buquerque | -12.769 |
WARE | -12.546 |
Startup | -12.189 |
Token | one |
Token ID | 530 |
Feature activation | +1.13e+01 |
Pos logit contributions | |
arthed | +4.335 |
##### | +4.180 |
ept | +4.124 |
Dol | +3.940 |
goodbye | +3.922 |
Neg logit contributions | |
ibaba | -4.801 |
racuse | -4.687 |
buquerque | -4.600 |
WARE | -4.592 |
Vector | -4.566 |
Token | is |
Token ID | 318 |
Feature activation | +7.88e+00 |
Pos logit contributions | |
arthed | +6.281 |
##### | +6.248 |
ept | +5.921 |
Dol | +5.899 |
goodbye | +5.767 |
Neg logit contributions | |
ibaba | -7.564 |
WARE | -7.367 |
racuse | -7.359 |
buquerque | -7.216 |
Vector | -6.965 |
Token | found |
Token ID | 1043 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
arthed | +5.211 |
##### | +4.855 |
ept | +4.848 |
Dol | +4.740 |
goodbye | +4.618 |
Neg logit contributions | |
ibaba | -5.090 |
racuse | -4.936 |
WARE | -4.782 |
buquerque | -4.782 |
Vector | -4.684 |
Token | to |
Token ID | 284 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
arthed | +1.815 |
ept | +1.707 |
##### | +1.680 |
iltr | +1.631 |
Dol | +1.602 |
Neg logit contributions | |
ibaba | -1.572 |
racuse | -1.495 |
Vector | -1.491 |
WARE | -1.466 |
buquerque | -1.451 |
Token | you |
Token ID | 345 |
Feature activation | +6.49e+00 |
Pos logit contributions | |
arthed | +4.556 |
goodbye | +4.415 |
Dol | +4.353 |
ept | +4.315 |
##### | +4.278 |
Neg logit contributions | |
ibaba | -6.295 |
WARE | -6.086 |
racuse | -6.043 |
Vector | -5.930 |
buquerque | -5.919 |
Token | should |
Token ID | 815 |
Feature activation | +9.16e+00 |
Pos logit contributions | |
##### | +3.886 |
arthed | +3.849 |
ept | +3.827 |
iltr | +3.706 |
goodbye | +3.684 |
Neg logit contributions | |
ibaba | -4.568 |
WARE | -4.410 |
racuse | -4.362 |
Vector | -4.299 |
buquerque | -4.225 |
Token | tell |
Token ID | 1560 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
##### | +4.822 |
goodbye | +4.676 |
ept | +4.545 |
Dol | +4.439 |
arthed | +4.424 |
Neg logit contributions | |
ibaba | -7.082 |
racuse | -6.915 |
WARE | -6.853 |
buquerque | -6.584 |
Vector | -6.567 |
Token | us |
Token ID | 514 |
Feature activation | +7.18e+00 |
Pos logit contributions | |
arthed | +0.211 |
ept | +0.198 |
##### | +0.193 |
iltr | +0.189 |
Dol | +0.186 |
Neg logit contributions | |
ibaba | -0.175 |
Vector | -0.167 |
racuse | -0.165 |
WARE | -0.162 |
buquerque | -0.161 |
Token | what |
Token ID | 644 |
Feature activation | +6.61e+00 |
Pos logit contributions | |
arthed | +4.142 |
goodbye | +4.046 |
##### | +3.951 |
ept | +3.932 |
Dol | +3.846 |
Neg logit contributions | |
ibaba | -5.325 |
racuse | -5.127 |
WARE | -5.017 |
Vector | -4.950 |
buquerque | -4.923 |
Token | to |
Token ID | 284 |
Feature activation | +2.02e+01 |
Pos logit contributions | |
arthed | +4.023 |
ept | +3.819 |
##### | +3.787 |
goodbye | +3.745 |
Dol | +3.722 |
Neg logit contributions | |
ibaba | -4.636 |
racuse | -4.466 |
WARE | -4.435 |
Vector | -4.433 |
buquerque | -4.340 |
Token | do |
Token ID | 466 |
Feature activation | +9.15e+00 |
Pos logit contributions | |
##### | +6.722 |
goodbye | +6.467 |
Dol | +6.270 |
iltr | +5.897 |
irs | +5.788 |
Neg logit contributions | |
racuse | -17.089 |
ibaba | -16.871 |
WARE | -16.357 |
buquerque | -15.857 |
Vector | -15.694 |
Token | ' |
Token ID | 6 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
arthed | +4.785 |
goodbye | +4.767 |
##### | +4.625 |
ept | +4.590 |
Dol | +4.482 |
Neg logit contributions | |
ibaba | -6.982 |
racuse | -6.720 |
WARE | -6.632 |
Vector | -6.516 |
buquerque | -6.422 |
Token |
|
Token ID | 198 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
ibaba | +0.623 |
Vector | +0.600 |
racuse | +0.587 |
WARE | +0.575 |
buquerque | +0.572 |
Neg logit contributions | |
arthed | -0.781 |
ept | -0.732 |
##### | -0.713 |
iltr | -0.703 |
Dol | -0.683 |
Token |
|
Token ID | 198 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
ibaba | +1.163 |
Vector | +1.151 |
racuse | +1.100 |
WARE | +1.086 |
ABLE | +1.077 |
Neg logit contributions | |
arthed | -1.651 |
ept | -1.545 |
iltr | -1.496 |
##### | -1.491 |
goodbye | -1.433 |
Token | We |
Token ID | 1135 |
Feature activation | +3.27e+00 |
Pos logit contributions | |
arthed | +0.287 |
ept | +0.269 |
##### | +0.263 |
iltr | +0.258 |
Dol | +0.253 |
Neg logit contributions | |
ibaba | -0.227 |
Vector | -0.217 |
racuse | -0.215 |
WARE | -0.209 |
buquerque | -0.208 |
Token | , |
Token ID | 11 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
arthed | +5.871 |
ept | +5.430 |
##### | +5.292 |
Dol | +5.273 |
goodbye | +5.231 |
Neg logit contributions | |
ibaba | -6.554 |
racuse | -6.332 |
buquerque | -6.083 |
WARE | -5.849 |
Vector | -5.795 |
Token | sub |
Token ID | 850 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
ibaba | +1.215 |
Vector | +1.179 |
racuse | +1.148 |
WARE | +1.127 |
buquerque | +1.110 |
Neg logit contributions | |
arthed | -1.574 |
ept | -1.482 |
##### | -1.445 |
iltr | -1.429 |
Dol | -1.378 |
Token | lim |
Token ID | 2475 |
Feature activation | -6.55e+00 |
Pos logit contributions | |
arthed | +0.139 |
ept | +0.129 |
##### | +0.126 |
iltr | +0.124 |
Dol | +0.121 |
Neg logit contributions | |
ibaba | -0.126 |
Vector | -0.121 |
racuse | -0.119 |
WARE | -0.117 |
buquerque | -0.116 |
Token | ed |
Token ID | 276 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
ibaba | +3.288 |
Vector | +3.027 |
racuse | +2.994 |
buquerque | +2.913 |
ABLE | +2.819 |
Neg logit contributions | |
arthed | -5.436 |
ept | -5.090 |
##### | -5.053 |
goodbye | -4.964 |
Dol | -4.893 |
Token | by |
Token ID | 416 |
Feature activation | +7.85e+00 |
Pos logit contributions | |
arthed | +1.896 |
ept | +1.768 |
##### | +1.727 |
iltr | +1.685 |
Dol | +1.672 |
Neg logit contributions | |
ibaba | -1.718 |
Vector | -1.633 |
racuse | -1.633 |
WARE | -1.598 |
buquerque | -1.589 |
Token | out |
Token ID | 503 |
Feature activation | +2.00e+01 |
Pos logit contributions | |
arthed | +5.616 |
ept | +5.216 |
##### | +5.168 |
Dol | +5.132 |
goodbye | +5.000 |
Neg logit contributions | |
ibaba | -4.807 |
racuse | -4.479 |
buquerque | -4.456 |
WARE | -4.440 |
Vector | -4.319 |
Token | g |
Token ID | 70 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
arthed | +11.238 |
bell | +10.606 |
ept | +10.414 |
##### | +10.163 |
Dol | +9.658 |
Neg logit contributions | |
ibaba | -12.282 |
racuse | -11.547 |
buquerque | -11.409 |
Notting | -11.360 |
Startup | -11.273 |
Token | assing |
Token ID | 19696 |
Feature activation | +7.17e+00 |
Pos logit contributions | |
ibaba | +0.937 |
Vector | +0.903 |
racuse | +0.892 |
WARE | +0.870 |
buquerque | +0.869 |
Neg logit contributions | |
arthed | -0.954 |
ept | -0.888 |
##### | -0.864 |
iltr | -0.843 |
Dol | -0.827 |
Token | , |
Token ID | 11 |
Feature activation | +9.36e-01 |
Pos logit contributions | |
arthed | +4.813 |
ept | +4.469 |
##### | +4.381 |
Dol | +4.332 |
iltr | +4.218 |
Neg logit contributions | |
ibaba | -4.764 |
racuse | -4.569 |
buquerque | -4.469 |
Vector | -4.353 |
WARE | -4.336 |
Token | or |
Token ID | 393 |
Feature activation | +8.49e+00 |
Pos logit contributions | |
arthed | +0.707 |
ept | +0.660 |
##### | +0.647 |
iltr | +0.632 |
Dol | +0.622 |
Neg logit contributions | |
ibaba | -0.562 |
Vector | -0.537 |
racuse | -0.531 |
WARE | -0.521 |
buquerque | -0.518 |
Token | simply |
Token ID | 2391 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
arthed | +5.849 |
ept | +5.318 |
##### | +5.312 |
Dol | +5.179 |
goodbye | +5.096 |
Neg logit contributions | |
ibaba | -5.379 |
buquerque | -5.152 |
racuse | -5.134 |
WARE | -4.999 |
Vector | -4.994 |
Token | are |
Token ID | 389 |
Feature activation | +7.38e+00 |
Pos logit contributions | |
arthed | +2.806 |
ept | +2.707 |
##### | +2.649 |
iltr | +2.591 |
Dol | +2.544 |
Neg logit contributions | |
ibaba | -2.591 |
racuse | -2.520 |
Vector | -2.456 |
WARE | -2.447 |
buquerque | -2.411 |
Token | the |
Token ID | 262 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
arthed | +4.824 |
ept | +4.580 |
##### | +4.576 |
Dol | +4.530 |
goodbye | +4.361 |
Neg logit contributions | |
ibaba | -4.778 |
racuse | -4.702 |
buquerque | -4.473 |
WARE | -4.412 |
Vector | -4.383 |
Token | greatest |
Token ID | 6000 |
Feature activation | +5.63e+00 |
Pos logit contributions | |
arthed | +1.426 |
ept | +1.344 |
##### | +1.322 |
iltr | +1.287 |
Dol | +1.278 |
Neg logit contributions | |
ibaba | -1.160 |
racuse | -1.099 |
Vector | -1.094 |
WARE | -1.070 |
buquerque | -1.064 |
Token | of |
Token ID | 286 |
Feature activation | +1.36e+01 |
Pos logit contributions | |
arthed | +3.722 |
ept | +3.523 |
##### | +3.480 |
Dol | +3.382 |
goodbye | +3.341 |
Neg logit contributions | |
ibaba | -3.781 |
racuse | -3.570 |
buquerque | -3.466 |
WARE | -3.458 |
Vector | -3.455 |
Token | the |
Token ID | 262 |
Feature activation | +1.65e+01 |
Pos logit contributions | |
arthed | +7.238 |
Dol | +6.929 |
ept | +6.808 |
##### | +6.788 |
goodbye | +6.613 |
Neg logit contributions | |
ibaba | -9.616 |
racuse | -9.458 |
WARE | -9.064 |
buquerque | -8.775 |
sidx | -8.594 |
Token | many |
Token ID | 867 |
Feature activation | +1.98e+01 |
Pos logit contributions | |
Dol | +8.220 |
##### | +8.207 |
arthed | +8.181 |
ept | +8.134 |
goodbye | +7.827 |
Neg logit contributions | |
ibaba | -11.295 |
racuse | -11.131 |
WARE | -10.595 |
buquerque | -10.441 |
_- | -10.004 |
Token | uncertainties |
Token ID | 36553 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
##### | +10.518 |
Dol | +10.414 |
goodbye | +10.104 |
ept | +10.066 |
arthed | +9.702 |
Neg logit contributions | |
ibaba | -12.217 |
racuse | -11.865 |
WARE | -11.158 |
buquerque | -11.115 |
sidx | -10.366 |
Token | that |
Token ID | 326 |
Feature activation | +6.30e+00 |
Pos logit contributions | |
arthed | +0.813 |
ept | +0.762 |
##### | +0.748 |
iltr | +0.729 |
Dol | +0.720 |
Neg logit contributions | |
ibaba | -0.712 |
racuse | -0.679 |
Vector | -0.676 |
WARE | -0.660 |
buquerque | -0.658 |
Token | complicate |
Token ID | 45058 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
arthed | +4.194 |
ept | +4.073 |
##### | +4.017 |
Dol | +3.928 |
iltr | +3.824 |
Neg logit contributions | |
ibaba | -4.004 |
racuse | -3.950 |
WARE | -3.736 |
buquerque | -3.710 |
Vector | -3.670 |
Token | the |
Token ID | 262 |
Feature activation | -2.78e-01 |
Pos logit contributions | |
ibaba | +0.686 |
Vector | +0.666 |
racuse | +0.648 |
WARE | +0.639 |
buquerque | +0.633 |
Neg logit contributions | |
arthed | -0.790 |
ept | -0.739 |
##### | -0.719 |
iltr | -0.709 |
Dol | -0.686 |
Token | determination |
Token ID | 12123 |
Feature activation | -7.16e+00 |
Pos logit contributions | |
ibaba | +0.176 |
Vector | +0.169 |
racuse | +0.167 |
WARE | +0.163 |
buquerque | +0.162 |
Neg logit contributions | |
arthed | -0.201 |
ept | -0.188 |
##### | -0.183 |
iltr | -0.180 |
Dol | -0.175 |
Token | ak |
Token ID | 461 |
Feature activation | +6.24e+00 |
Pos logit contributions | |
arthed | +1.403 |
ept | +1.305 |
##### | +1.281 |
iltr | +1.254 |
Dol | +1.231 |
Neg logit contributions | |
ibaba | -1.226 |
Vector | -1.185 |
racuse | -1.182 |
WARE | -1.158 |
buquerque | -1.147 |
Token | by |
Token ID | 416 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
arthed | +4.320 |
ept | +4.076 |
##### | +4.007 |
Dol | +3.914 |
iltr | +3.859 |
Neg logit contributions | |
ibaba | -3.894 |
racuse | -3.802 |
WARE | -3.697 |
buquerque | -3.637 |
Vector | -3.632 |
Token | blocking |
Token ID | 12013 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
arthed | +1.031 |
ept | +0.966 |
##### | +0.950 |
iltr | +0.926 |
Dol | +0.913 |
Neg logit contributions | |
ibaba | -0.889 |
racuse | -0.846 |
Vector | -0.844 |
WARE | -0.824 |
buquerque | -0.821 |
Token | national |
Token ID | 2260 |
Feature activation | +6.30e+00 |
Pos logit contributions | |
arthed | +1.841 |
ept | +1.700 |
##### | +1.682 |
iltr | +1.629 |
Dol | +1.620 |
Neg logit contributions | |
ibaba | -1.701 |
racuse | -1.646 |
Vector | -1.618 |
buquerque | -1.590 |
WARE | -1.588 |
Token | highway |
Token ID | 12763 |
Feature activation | +1.07e+01 |
Pos logit contributions | |
arthed | +3.997 |
ept | +3.741 |
##### | +3.675 |
goodbye | +3.631 |
Dol | +3.593 |
Neg logit contributions | |
ibaba | -4.218 |
racuse | -4.182 |
buquerque | -4.038 |
WARE | -4.028 |
Vector | -3.981 |
Token | connecting |
Token ID | 14320 |
Feature activation | +1.98e+01 |
Pos logit contributions | |
arthed | +6.710 |
##### | +6.590 |
ept | +6.316 |
Dol | +6.253 |
goodbye | +6.189 |
Neg logit contributions | |
racuse | -7.147 |
ibaba | -6.880 |
buquerque | -6.576 |
WARE | -6.486 |
ABLE | -6.281 |
Token | Roh |
Token ID | 32694 |
Feature activation | +9.07e+00 |
Pos logit contributions | |
Dol | +12.468 |
arthed | +12.055 |
##### | +11.976 |
ept | +11.685 |
ⓘ | +11.295 |
Neg logit contributions | |
racuse | -10.331 |
ibaba | -9.713 |
WARE | -9.704 |
ABLE | -9.221 |
buquerque | -9.212 |
Token | t |
Token ID | 83 |
Feature activation | -9.48e-02 |
Pos logit contributions | |
arthed | +3.926 |
ept | +3.625 |
Dol | +3.361 |
iltr | +3.310 |
##### | +3.283 |
Neg logit contributions | |
WARE | -7.790 |
racuse | -7.713 |
ibaba | -7.675 |
ABLE | -7.637 |
Vector | -7.517 |
Token | ak |
Token ID | 461 |
Feature activation | +7.65e+00 |
Pos logit contributions | |
ibaba | +0.059 |
Vector | +0.057 |
racuse | +0.056 |
WARE | +0.055 |
buquerque | +0.054 |
Neg logit contributions | |
arthed | -0.069 |
ept | -0.065 |
##### | -0.063 |
iltr | -0.062 |
Dol | -0.061 |
Token | to |
Token ID | 284 |
Feature activation | +9.43e+00 |
Pos logit contributions | |
arthed | +4.793 |
ept | +4.566 |
##### | +4.549 |
Dol | +4.413 |
iltr | +4.346 |
Neg logit contributions | |
ibaba | -5.096 |
racuse | -4.970 |
WARE | -4.892 |
buquerque | -4.770 |
Vector | -4.769 |
Token | Delhi |
Token ID | 12517 |
Feature activation | +4.57e+00 |
Pos logit contributions | |
arthed | +6.664 |
Dol | +6.466 |
ept | +6.414 |
##### | +6.370 |
ⓘ | +6.079 |
Neg logit contributions | |
racuse | -5.356 |
ibaba | -5.270 |
WARE | -5.156 |
buquerque | -4.970 |
ABLE | -4.913 |
Token | House |
Token ID | 2097 |
Feature activation | -4.53e+00 |
Pos logit contributions | |
Vector | +2.022 |
ibaba | +1.998 |
Higgins | +1.914 |
racuse | +1.910 |
Startup | +1.896 |
Neg logit contributions | |
arthed | -3.138 |
ept | -2.942 |
##### | -2.831 |
iltr | -2.828 |
Dol | -2.717 |
Token | speaker |
Token ID | 10834 |
Feature activation | -7.45e-03 |
Pos logit contributions | |
Vector | +2.172 |
ibaba | +2.148 |
Higgins | +2.029 |
racuse | +2.009 |
WARE | +2.007 |
Neg logit contributions | |
arthed | -3.889 |
ept | -3.653 |
iltr | -3.492 |
##### | -3.477 |
Dol | -3.377 |
Token | whose |
Token ID | 3025 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
ibaba | +0.005 |
Vector | +0.004 |
racuse | +0.004 |
WARE | +0.004 |
buquerque | +0.004 |
Neg logit contributions | |
arthed | -0.005 |
ept | -0.005 |
##### | -0.005 |
iltr | -0.005 |
Dol | -0.005 |
Token | conservative |
Token ID | 5940 |
Feature activation | +5.57e+00 |
Pos logit contributions | |
ibaba | +1.086 |
Vector | +1.064 |
racuse | +1.030 |
WARE | +1.017 |
Higgins | +1.005 |
Neg logit contributions | |
arthed | -1.371 |
ept | -1.283 |
##### | -1.243 |
iltr | -1.230 |
Dol | -1.188 |
Token | positions |
Token ID | 6116 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
arthed | +3.925 |
##### | +3.678 |
ept | +3.673 |
Dol | +3.585 |
goodbye | +3.541 |
Neg logit contributions | |
ibaba | -3.543 |
racuse | -3.364 |
Vector | -3.231 |
buquerque | -3.230 |
WARE | -3.160 |
Token | on |
Token ID | 319 |
Feature activation | +1.97e+01 |
Pos logit contributions | |
arthed | +1.731 |
ept | +1.625 |
##### | +1.601 |
iltr | +1.563 |
Dol | +1.552 |
Neg logit contributions | |
ibaba | -1.614 |
racuse | -1.534 |
Vector | -1.510 |
WARE | -1.489 |
buquerque | -1.479 |
Token | immigration |
Token ID | 6272 |
Feature activation | +6.84e+00 |
Pos logit contributions | |
arthed | +9.182 |
##### | +9.172 |
Dol | +8.991 |
goodbye | +8.774 |
ept | +8.232 |
Neg logit contributions | |
ibaba | -13.787 |
racuse | -13.254 |
WARE | -12.642 |
buquerque | -12.381 |
sidx | -12.328 |
Token | were |
Token ID | 547 |
Feature activation | +7.45e-01 |
Pos logit contributions | |
arthed | +4.418 |
ept | +4.218 |
##### | +4.178 |
iltr | +4.082 |
goodbye | +4.049 |
Neg logit contributions | |
ibaba | -4.615 |
racuse | -4.425 |
WARE | -4.251 |
Vector | -4.236 |
buquerque | -4.146 |
Token | far |
Token ID | 1290 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
arthed | +0.540 |
ept | +0.504 |
##### | +0.493 |
iltr | +0.483 |
Dol | +0.476 |
Neg logit contributions | |
ibaba | -0.470 |
Vector | -0.448 |
racuse | -0.444 |
WARE | -0.435 |
buquerque | -0.433 |
Token | too |
Token ID | 1165 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
arthed | +1.513 |
ept | +1.403 |
##### | +1.380 |
iltr | +1.351 |
Dol | +1.334 |
Neg logit contributions | |
ibaba | -1.358 |
Vector | -1.295 |
racuse | -1.271 |
buquerque | -1.245 |
WARE | -1.242 |
Token | per |
Token ID | 583 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
arthed | +1.383 |
ept | +1.288 |
##### | +1.259 |
iltr | +1.231 |
Dol | +1.228 |
Neg logit contributions | |
ibaba | -1.195 |
racuse | -1.128 |
Vector | -1.128 |
buquerque | -1.097 |
WARE | -1.097 |
Token | C |
Token ID | 327 |
Feature activation | -1.07e+01 |
Pos logit contributions | |
arthed | +5.346 |
##### | +4.970 |
ept | +4.969 |
Dol | +4.934 |
goodbye | +4.767 |
Neg logit contributions | |
ibaba | -4.075 |
racuse | -3.785 |
WARE | -3.692 |
buquerque | -3.645 |
Vector | -3.626 |
Token | NET |
Token ID | 12884 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
ABLE | +4.753 |
ibaba | +4.669 |
Vector | +4.403 |
ables | +4.366 |
racuse | +4.333 |
Neg logit contributions | |
arthed | -9.024 |
##### | -8.438 |
iltr | -8.403 |
goodbye | -8.243 |
Dol | -8.193 |
Token | 's |
Token ID | 338 |
Feature activation | +1.34e+01 |
Pos logit contributions | |
arthed | +1.958 |
ept | +1.839 |
##### | +1.799 |
iltr | +1.738 |
goodbye | +1.732 |
Neg logit contributions | |
ibaba | -1.547 |
Vector | -1.465 |
racuse | -1.458 |
WARE | -1.426 |
buquerque | -1.408 |
Token | in |
Token ID | 287 |
Feature activation | +8.32e+00 |
Pos logit contributions | |
arthed | +8.708 |
Dol | +8.357 |
goodbye | +8.262 |
ept | +8.208 |
##### | +8.199 |
Neg logit contributions | |
ibaba | -7.792 |
racuse | -7.551 |
WARE | -7.499 |
buquerque | -7.310 |
ABLE | -7.177 |
Token | - |
Token ID | 12 |
Feature activation | +6.69e+00 |
Pos logit contributions | |
arthed | +5.318 |
ept | +4.993 |
##### | +4.821 |
Dol | +4.782 |
iltr | +4.551 |
Neg logit contributions | |
ibaba | -5.677 |
WARE | -5.323 |
racuse | -5.289 |
Vector | -5.279 |
buquerque | -5.141 |
Token | depth |
Token ID | 18053 |
Feature activation | +1.96e+01 |
Pos logit contributions | |
arthed | +2.951 |
ept | +2.734 |
##### | +2.540 |
bell | +2.417 |
Dol | +2.410 |
Neg logit contributions | |
ibaba | -5.968 |
Vector | -5.751 |
racuse | -5.699 |
WARE | -5.668 |
Notting | -5.550 |
Token | coverage |
Token ID | 5197 |
Feature activation | +6.73e+00 |
Pos logit contributions | |
Dol | +10.268 |
goodbye | +10.125 |
arthed | +9.987 |
ept | +9.575 |
##### | +9.517 |
Neg logit contributions | |
ibaba | -11.933 |
racuse | -11.648 |
WARE | -11.513 |
ABLE | -11.144 |
buquerque | -11.141 |
Token | . |
Token ID | 13 |
Feature activation | +9.99e-01 |
Pos logit contributions | |
arthed | +4.323 |
ept | +3.984 |
##### | +3.941 |
Dol | +3.861 |
goodbye | +3.854 |
Neg logit contributions | |
ibaba | -4.605 |
racuse | -4.383 |
Vector | -4.282 |
WARE | -4.265 |
buquerque | -4.234 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
arthed | +0.778 |
ept | +0.730 |
##### | +0.719 |
iltr | +0.702 |
Dol | +0.694 |
Neg logit contributions | |
ibaba | -0.573 |
Vector | -0.544 |
racuse | -0.539 |
WARE | -0.527 |
buquerque | -0.523 |
Token | After |
Token ID | 3260 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
arthed | +1.900 |
ept | +1.781 |
##### | +1.730 |
iltr | +1.699 |
Dol | +1.656 |
Neg logit contributions | |
ibaba | -1.701 |
Vector | -1.638 |
racuse | -1.621 |
buquerque | -1.566 |
WARE | -1.548 |
Token | Saturday |
Token ID | 3909 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
arthed | +1.201 |
ept | +1.125 |
##### | +1.099 |
iltr | +1.074 |
Dol | +1.061 |
Neg logit contributions | |
ibaba | -1.064 |
Vector | -1.019 |
racuse | -1.016 |
WARE | -0.992 |
buquerque | -0.982 |
Token | in |
Token ID | 287 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
arthed | +0.982 |
ept | +0.923 |
##### | +0.906 |
iltr | +0.881 |
goodbye | +0.872 |
Neg logit contributions | |
ibaba | -0.857 |
Vector | -0.817 |
racuse | -0.813 |
WARE | -0.797 |
buquerque | -0.785 |
Token | 4 |
Token ID | 604 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
ibaba | +1.361 |
Vector | +1.324 |
racuse | +1.281 |
WARE | +1.251 |
Higgins | +1.251 |
Neg logit contributions | |
arthed | -1.806 |
ept | -1.681 |
##### | -1.632 |
iltr | -1.628 |
Dol | -1.563 |
Token | hours |
Token ID | 2250 |
Feature activation | -3.73e+00 |
Pos logit contributions | |
ibaba | +0.126 |
Vector | +0.122 |
racuse | +0.120 |
WARE | +0.118 |
buquerque | +0.117 |
Neg logit contributions | |
arthed | -0.117 |
ept | -0.108 |
##### | -0.106 |
iltr | -0.103 |
Dol | -0.101 |
Token | . |
Token ID | 13 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
ibaba | +2.026 |
Vector | +1.969 |
racuse | +1.885 |
WARE | +1.856 |
buquerque | +1.855 |
Neg logit contributions | |
arthed | -3.024 |
ept | -2.823 |
iltr | -2.722 |
##### | -2.712 |
Dol | -2.627 |
Token | That |
Token ID | 1320 |
Feature activation | +5.73e-02 |
Pos logit contributions | |
arthed | +1.265 |
ept | +1.185 |
##### | +1.179 |
iltr | +1.136 |
Dol | +1.135 |
Neg logit contributions | |
ibaba | -1.075 |
racuse | -1.021 |
Vector | -1.008 |
WARE | -0.990 |
buquerque | -0.987 |
Token | must |
Token ID | 1276 |
Feature activation | +1.96e+01 |
Pos logit contributions | |
arthed | +0.044 |
ept | +0.042 |
##### | +0.041 |
iltr | +0.040 |
Dol | +0.039 |
Neg logit contributions | |
ibaba | -0.033 |
Vector | -0.032 |
racuse | -0.031 |
WARE | -0.030 |
buquerque | -0.030 |
Token | be |
Token ID | 307 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
##### | +8.632 |
ept | +8.442 |
auna | +8.337 |
bell | +8.064 |
arthed | +8.044 |
Neg logit contributions | |
ibaba | -13.503 |
WARE | -13.273 |
Vector | -13.040 |
racuse | -13.007 |
ABLE | -12.612 |
Token | a |
Token ID | 257 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
arthed | +1.172 |
ept | +1.099 |
##### | +1.079 |
iltr | +1.046 |
Dol | +1.040 |
Neg logit contributions | |
ibaba | -1.098 |
Vector | -1.053 |
racuse | -1.049 |
WARE | -1.027 |
buquerque | -1.015 |
Token | new |
Token ID | 649 |
Feature activation | -8.92e-01 |
Pos logit contributions | |
ibaba | +0.920 |
Vector | +0.889 |
racuse | +0.866 |
WARE | +0.851 |
buquerque | +0.847 |
Neg logit contributions | |
arthed | -1.134 |
ept | -1.051 |
##### | -1.024 |
iltr | -1.015 |
Dol | -0.979 |
Token | record |
Token ID | 1700 |
Feature activation | +7.56e-01 |
Pos logit contributions | |
ibaba | +0.543 |
Vector | +0.523 |
racuse | +0.513 |
WARE | +0.502 |
buquerque | +0.501 |
Neg logit contributions | |
arthed | -0.664 |
ept | -0.620 |
##### | -0.606 |
iltr | -0.595 |
Dol | -0.580 |
Token | . |
Token ID | 13 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
arthed | +0.547 |
ept | +0.514 |
##### | +0.502 |
iltr | +0.492 |
Dol | +0.482 |
Neg logit contributions | |
ibaba | -0.473 |
Vector | -0.455 |
racuse | -0.450 |
WARE | -0.440 |
buquerque | -0.436 |
Token | supporting |
Token ID | 6493 |
Feature activation | +6.85e+00 |
Pos logit contributions | |
ibaba | +1.531 |
Vector | +1.500 |
racuse | +1.454 |
WARE | +1.427 |
buquerque | +1.413 |
Neg logit contributions | |
arthed | -1.981 |
ept | -1.853 |
##### | -1.807 |
iltr | -1.787 |
Dol | -1.719 |
Token | edible |
Token ID | 35988 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
arthed | +4.413 |
ept | +4.206 |
##### | +4.193 |
Dol | +4.174 |
goodbye | +4.064 |
Neg logit contributions | |
ibaba | -4.642 |
racuse | -4.402 |
WARE | -4.279 |
buquerque | -4.201 |
Vector | -4.131 |
Token | plants |
Token ID | 6134 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
arthed | +0.008 |
ept | +0.008 |
##### | +0.008 |
iltr | +0.007 |
Dol | +0.007 |
Neg logit contributions | |
ibaba | -0.008 |
Vector | -0.008 |
racuse | -0.008 |
WARE | -0.007 |
buquerque | -0.007 |
Token | that |
Token ID | 326 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
ibaba | +0.679 |
Vector | +0.659 |
racuse | +0.643 |
WARE | +0.631 |
buquerque | +0.625 |
Neg logit contributions | |
arthed | -0.839 |
ept | -0.783 |
##### | -0.764 |
iltr | -0.751 |
Dol | -0.731 |
Token | require |
Token ID | 2421 |
Feature activation | +1.20e+01 |
Pos logit contributions | |
arthed | +0.078 |
ept | +0.073 |
##### | +0.071 |
iltr | +0.070 |
Dol | +0.068 |
Neg logit contributions | |
ibaba | -0.066 |
Vector | -0.063 |
racuse | -0.062 |
WARE | -0.061 |
buquerque | -0.060 |
Token | little |
Token ID | 1310 |
Feature activation | +1.91e+01 |
Pos logit contributions | |
arthed | +7.465 |
ept | +7.299 |
Dol | +7.147 |
##### | +7.142 |
goodbye | +7.002 |
Neg logit contributions | |
ibaba | -7.610 |
racuse | -7.369 |
buquerque | -7.202 |
sidx | -6.880 |
WARE | -6.879 |
Token | to |
Token ID | 284 |
Feature activation | +1.32e+01 |
Pos logit contributions | |
##### | +9.895 |
ept | +9.828 |
arthed | +9.816 |
Dol | +9.638 |
goodbye | +9.379 |
Neg logit contributions | |
racuse | -11.922 |
ibaba | -11.669 |
buquerque | -11.438 |
Notting | -10.804 |
ABLE | -10.802 |
Token | no |
Token ID | 645 |
Feature activation | +1.87e+01 |
Pos logit contributions | |
ept | +7.241 |
arthed | +7.220 |
##### | +6.944 |
goodbye | +6.801 |
Dol | +6.630 |
Neg logit contributions | |
racuse | -9.084 |
ibaba | -8.612 |
buquerque | -8.612 |
WARE | -8.484 |
Vector | -8.448 |
Token | further |
Token ID | 2252 |
Feature activation | +1.53e+01 |
Pos logit contributions | |
arthed | +9.828 |
##### | +9.757 |
ept | +9.655 |
Dol | +9.587 |
goodbye | +9.392 |
Neg logit contributions | |
racuse | -11.636 |
ibaba | -11.360 |
buquerque | -10.946 |
Notting | -10.729 |
sidx | -10.532 |
Token | gardening |
Token ID | 46072 |
Feature activation | +1.45e+01 |
Pos logit contributions | |
arthed | +8.809 |
ept | +8.485 |
##### | +8.356 |
Dol | +8.314 |
iltr | +8.141 |
Neg logit contributions | |
ibaba | -9.680 |
racuse | -9.680 |
buquerque | -9.107 |
Notting | -8.811 |
WARE | -8.801 |
Token | . |
Token ID | 13 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
arthed | +7.680 |
ept | +7.655 |
##### | +7.391 |
Dol | +7.179 |
iltr | +6.985 |
Neg logit contributions | |
ibaba | -10.209 |
racuse | -10.024 |
buquerque | -9.528 |
Higgins | -9.127 |
ABLE | -9.028 |
Token | bl |
Token ID | 698 |
Feature activation | +7.80e+00 |
Pos logit contributions | |
arthed | +8.655 |
##### | +8.455 |
Dol | +8.364 |
ept | +8.211 |
iltr | +8.170 |
Neg logit contributions | |
ibaba | -8.301 |
racuse | -8.004 |
buquerque | -7.506 |
sidx | -7.339 |
Higgins | -7.094 |
Token | urring |
Token ID | 14924 |
Feature activation | +1.79e+01 |
Pos logit contributions | |
arthed | +4.548 |
##### | +4.255 |
ept | +4.244 |
iltr | +4.237 |
irs | +4.011 |
Neg logit contributions | |
ibaba | -5.509 |
Vector | -5.509 |
WARE | -5.474 |
racuse | -5.337 |
ABLE | -5.309 |
Token | is |
Token ID | 318 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
##### | +9.941 |
Dol | +9.431 |
goodbye | +9.391 |
arthed | +9.379 |
ept | +8.956 |
Neg logit contributions | |
racuse | -11.559 |
ibaba | -11.558 |
buquerque | -10.249 |
sidx | -10.232 |
WARE | -10.013 |
Token | applied |
Token ID | 5625 |
Feature activation | +9.09e+00 |
Pos logit contributions | |
arthed | +6.359 |
##### | +6.279 |
Dol | +6.263 |
ept | +6.171 |
goodbye | +6.161 |
Neg logit contributions | |
racuse | -6.977 |
ibaba | -6.929 |
sidx | -6.473 |
buquerque | -6.473 |
WARE | -6.317 |
Token | on |
Token ID | 319 |
Feature activation | +8.94e+00 |
Pos logit contributions | |
arthed | +5.315 |
ept | +5.202 |
##### | +5.091 |
goodbye | +4.950 |
Dol | +4.946 |
Neg logit contributions | |
ibaba | -6.491 |
racuse | -6.208 |
buquerque | -6.092 |
WARE | -5.928 |
Vector | -5.847 |
Token | Line |
Token ID | 6910 |
Feature activation | +1.91e+01 |
Pos logit contributions | |
arthed | +5.510 |
##### | +5.287 |
ept | +5.181 |
Dol | +5.170 |
goodbye | +4.986 |
Neg logit contributions | |
ibaba | -5.994 |
racuse | -5.886 |
buquerque | -5.628 |
WARE | -5.623 |
Vector | -5.494 |
Token | 26 |
Token ID | 2608 |
Feature activation | +9.90e+00 |
Pos logit contributions | |
##### | +9.501 |
Dol | +8.573 |
arthed | +8.475 |
goodbye | +8.223 |
ept | +8.108 |
Neg logit contributions | |
ibaba | -13.576 |
racuse | -13.325 |
buquerque | -12.720 |
WARE | -12.596 |
ccording | -12.316 |
Token | to |
Token ID | 284 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
arthed | +5.991 |
##### | +5.876 |
ept | +5.719 |
goodbye | +5.593 |
Dol | +5.476 |
Neg logit contributions | |
ibaba | -6.633 |
racuse | -6.468 |
buquerque | -6.192 |
Vector | -6.142 |
WARE | -6.090 |
Token | reduce |
Token ID | 4646 |
Feature activation | +1.19e+01 |
Pos logit contributions | |
##### | +6.834 |
arthed | +6.830 |
ept | +6.696 |
Dol | +6.597 |
goodbye | +6.370 |
Neg logit contributions | |
racuse | -6.847 |
ibaba | -6.838 |
buquerque | -6.461 |
sidx | -6.204 |
WARE | -6.189 |
Token | high |
Token ID | 1029 |
Feature activation | +5.70e+00 |
Pos logit contributions | |
arthed | +6.727 |
##### | +6.655 |
Dol | +6.583 |
ept | +6.312 |
goodbye | +6.225 |
Neg logit contributions | |
ibaba | -7.947 |
racuse | -7.856 |
buquerque | -7.430 |
WARE | -7.422 |
sidx | -7.299 |
Token | frequency |
Token ID | 8373 |
Feature activation | +9.32e+00 |
Pos logit contributions | |
arthed | +3.767 |
ept | +3.629 |
##### | +3.604 |
iltr | +3.465 |
Dol | +3.461 |
Neg logit contributions | |
ibaba | -3.688 |
racuse | -3.567 |
Vector | -3.467 |
WARE | -3.403 |
buquerque | -3.383 |
Token | R |
Token ID | 49 |
Feature activation | +4.85e+00 |
Pos logit contributions | |
arthed | +3.065 |
ept | +2.883 |
##### | +2.833 |
iltr | +2.748 |
Dol | +2.700 |
Neg logit contributions | |
ibaba | -2.651 |
racuse | -2.550 |
Vector | -2.500 |
buquerque | -2.435 |
Higgins | -2.366 |
Token | S |
Token ID | 311 |
Feature activation | +9.94e+00 |
Pos logit contributions | |
arthed | +3.391 |
ept | +3.200 |
##### | +3.140 |
iltr | +3.059 |
Dol | +3.024 |
Neg logit contributions | |
ibaba | -3.032 |
racuse | -2.995 |
Vector | -2.920 |
WARE | -2.902 |
buquerque | -2.821 |
Token | enth |
Token ID | 7944 |
Feature activation | +6.91e+00 |
Pos logit contributions | |
arthed | +4.146 |
ept | +3.998 |
iltr | +3.614 |
##### | +3.508 |
irs | +3.269 |
Neg logit contributions | |
WARE | -8.578 |
Vector | -8.548 |
racuse | -8.504 |
buquerque | -8.463 |
ibaba | -8.446 |
Token | il |
Token ID | 346 |
Feature activation | -7.90e-02 |
Pos logit contributions | |
arthed | +2.061 |
ept | +1.899 |
iltr | +1.797 |
##### | +1.708 |
Dol | +1.605 |
Neg logit contributions | |
WARE | -7.004 |
racuse | -6.859 |
ibaba | -6.854 |
buquerque | -6.796 |
Vector | -6.781 |
Token | Kumar |
Token ID | 26105 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
ibaba | +0.032 |
Vector | +0.030 |
racuse | +0.029 |
WARE | +0.029 |
buquerque | +0.028 |
Neg logit contributions | |
arthed | -0.075 |
ept | -0.071 |
##### | -0.070 |
iltr | -0.069 |
Dol | -0.068 |
Token | / |
Token ID | 14 |
Feature activation | +1.91e+01 |
Pos logit contributions | |
arthed | +2.693 |
ept | +2.546 |
##### | +2.542 |
iltr | +2.441 |
Dol | +2.418 |
Neg logit contributions | |
ibaba | -2.187 |
racuse | -2.105 |
Vector | -2.066 |
WARE | -2.037 |
buquerque | -2.008 |
Token | Press |
Token ID | 4332 |
Feature activation | +1.20e+01 |
Pos logit contributions | |
arthed | +9.619 |
##### | +9.581 |
ept | +9.565 |
bell | +9.064 |
CTV | +8.754 |
Neg logit contributions | |
racuse | -12.497 |
ibaba | -12.323 |
sidx | -11.298 |
Vector | -11.193 |
buquerque | -11.097 |
Token | Trust |
Token ID | 9870 |
Feature activation | +5.19e+00 |
Pos logit contributions | |
##### | +7.690 |
arthed | +7.229 |
ept | +7.210 |
Dol | +6.923 |
bell | +6.815 |
Neg logit contributions | |
ibaba | -8.055 |
racuse | -7.763 |
ccording | -7.695 |
buquerque | -7.424 |
WARE | -7.245 |
Token | of |
Token ID | 286 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
arthed | +3.197 |
##### | +3.064 |
ept | +3.044 |
Dol | +2.934 |
goodbye | +2.817 |
Neg logit contributions | |
ibaba | -3.723 |
racuse | -3.599 |
buquerque | -3.513 |
WARE | -3.458 |
Vector | -3.380 |
Token | India |
Token ID | 3794 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
arthed | +0.186 |
ept | +0.176 |
##### | +0.173 |
iltr | +0.170 |
Dol | +0.167 |
Neg logit contributions | |
ibaba | -0.103 |
Vector | -0.098 |
racuse | -0.096 |
WARE | -0.094 |
buquerque | -0.093 |
Token | via |
Token ID | 2884 |
Feature activation | +1.30e+01 |
Pos logit contributions | |
arthed | +4.042 |
##### | +3.897 |
ept | +3.656 |
Dol | +3.561 |
/# | +3.342 |
Neg logit contributions | |
racuse | -9.479 |
ibaba | -9.427 |
WARE | -8.965 |
buquerque | -8.930 |
Vector | -8.876 |
Token | M |
Token ID | 337 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
arthed | +5.996 |
##### | +5.539 |
Dol | +5.525 |
ept | +5.481 |
goodbye | +5.311 |
Neg logit contributions | |
ibaba | -5.889 |
racuse | -5.533 |
WARE | -5.455 |
buquerque | -5.346 |
Vector | -5.274 |
Token | WC |
Token ID | 27353 |
Feature activation | +5.77e+00 |
Pos logit contributions | |
arthed | +2.463 |
ept | +2.297 |
##### | +2.175 |
iltr | +2.165 |
Dol | +2.057 |
Neg logit contributions | |
ibaba | -3.169 |
Vector | -3.103 |
racuse | -3.048 |
buquerque | -2.997 |
WARE | -2.981 |
Token | 12 |
Token ID | 1105 |
Feature activation | +6.70e+00 |
Pos logit contributions | |
arthed | +3.752 |
ept | +3.476 |
##### | +3.420 |
Dol | +3.317 |
goodbye | +3.314 |
Neg logit contributions | |
ibaba | -3.911 |
racuse | -3.763 |
Vector | -3.745 |
WARE | -3.685 |
buquerque | -3.669 |
Token | as |
Token ID | 355 |
Feature activation | +1.27e+01 |
Pos logit contributions | |
arthed | +4.245 |
ept | +4.089 |
##### | +3.982 |
Dol | +3.756 |
iltr | +3.724 |
Neg logit contributions | |
ibaba | -4.587 |
racuse | -4.456 |
Vector | -4.355 |
WARE | -4.340 |
buquerque | -4.280 |
Token | a |
Token ID | 257 |
Feature activation | +1.19e+01 |
Pos logit contributions | |
arthed | +7.291 |
ept | +6.794 |
##### | +6.703 |
goodbye | +6.619 |
Dol | +6.555 |
Neg logit contributions | |
ibaba | -8.655 |
racuse | -8.341 |
WARE | -8.210 |
Vector | -7.977 |
buquerque | -7.894 |
Token | legit |
Token ID | 6984 |
Feature activation | +1.90e+01 |
Pos logit contributions | |
arthed | +7.452 |
Dol | +7.222 |
ept | +7.164 |
##### | +7.160 |
goodbye | +7.120 |
Neg logit contributions | |
ibaba | -7.455 |
racuse | -7.248 |
WARE | -7.130 |
buquerque | -6.990 |
heit | -6.789 |
Token | B |
Token ID | 347 |
Feature activation | +8.76e+00 |
Pos logit contributions | |
arthed | +10.452 |
goodbye | +10.235 |
Dol | +10.232 |
ept | +9.972 |
##### | +9.936 |
Neg logit contributions | |
ibaba | -11.337 |
WARE | -11.121 |
racuse | -10.930 |
buquerque | -10.559 |
heit | -10.474 |
Token | CS |
Token ID | 7902 |
Feature activation | +1.47e+01 |
Pos logit contributions | |
arthed | +3.109 |
ept | +2.845 |
##### | +2.622 |
iltr | +2.603 |
CTV | +2.422 |
Neg logit contributions | |
ibaba | -8.301 |
Vector | -8.103 |
racuse | -8.097 |
buquerque | -7.970 |
WARE | -7.936 |
Token | conference |
Token ID | 4495 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
arthed | +8.872 |
ept | +8.073 |
Dol | +7.980 |
##### | +7.881 |
goodbye | +7.870 |
Neg logit contributions | |
ibaba | -9.470 |
WARE | -8.858 |
racuse | -8.738 |
buquerque | -8.538 |
Vector | -8.506 |
Token | . |
Token ID | 13 |
Feature activation | +6.06e+00 |
Pos logit contributions | |
arthed | +6.280 |
ept | +5.775 |
goodbye | +5.643 |
Dol | +5.551 |
##### | +5.541 |
Neg logit contributions | |
ibaba | -7.207 |
WARE | -6.898 |
racuse | -6.882 |
buquerque | -6.748 |
Vector | -6.710 |
Token |
|
Token ID | 198 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
arthed | +3.921 |
##### | +3.767 |
Dol | +3.657 |
ept | +3.643 |
goodbye | +3.456 |
Neg logit contributions | |
ibaba | -4.192 |
racuse | -3.923 |
WARE | -3.821 |
buquerque | -3.766 |
Vector | -3.713 |
Token | third |
Token ID | 2368 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
arthed | +5.943 |
ept | +5.900 |
##### | +5.858 |
Dol | +5.783 |
goodbye | +5.613 |
Neg logit contributions | |
ibaba | -7.018 |
WARE | -6.759 |
buquerque | -6.642 |
racuse | -6.526 |
ABLE | -6.384 |
Token | overall |
Token ID | 4045 |
Feature activation | +3.52e+00 |
Pos logit contributions | |
arthed | +1.043 |
ept | +0.989 |
##### | +0.966 |
iltr | +0.946 |
Dol | +0.932 |
Neg logit contributions | |
ibaba | -0.792 |
Vector | -0.752 |
racuse | -0.742 |
WARE | -0.735 |
buquerque | -0.724 |
Token | pick |
Token ID | 2298 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
arthed | +2.560 |
ept | +2.447 |
##### | +2.378 |
iltr | +2.328 |
Dol | +2.291 |
Neg logit contributions | |
ibaba | -2.125 |
Vector | -2.031 |
WARE | -2.007 |
racuse | -2.006 |
buquerque | -1.974 |
Token | in |
Token ID | 287 |
Feature activation | +3.39e+00 |
Pos logit contributions | |
arthed | +1.444 |
ept | +1.355 |
##### | +1.326 |
iltr | +1.293 |
goodbye | +1.272 |
Neg logit contributions | |
ibaba | -1.336 |
Vector | -1.279 |
racuse | -1.268 |
WARE | -1.248 |
buquerque | -1.231 |
Token | the |
Token ID | 262 |
Feature activation | +7.71e+00 |
Pos logit contributions | |
arthed | +2.334 |
ept | +2.194 |
##### | +2.132 |
Dol | +2.068 |
iltr | +2.064 |
Neg logit contributions | |
ibaba | -2.196 |
Vector | -2.132 |
racuse | -2.099 |
WARE | -2.095 |
buquerque | -2.042 |
Token | 2015 |
Token ID | 1853 |
Feature activation | +1.90e+01 |
Pos logit contributions | |
arthed | +4.745 |
ept | +4.603 |
##### | +4.527 |
Dol | +4.407 |
goodbye | +4.267 |
Neg logit contributions | |
ibaba | -5.211 |
WARE | -5.064 |
racuse | -5.004 |
Vector | -4.887 |
buquerque | -4.887 |
Token | NBA |
Token ID | 7403 |
Feature activation | +1.34e+01 |
Pos logit contributions | |
Dol | +8.218 |
ept | +8.190 |
##### | +7.877 |
arthed | +7.780 |
/# | +7.463 |
Neg logit contributions | |
WARE | -13.612 |
ibaba | -13.494 |
racuse | -12.852 |
buquerque | -12.704 |
heit | -12.435 |
Token | draft |
Token ID | 4538 |
Feature activation | +8.12e+00 |
Pos logit contributions | |
arthed | +7.101 |
ept | +6.670 |
##### | +6.360 |
Dol | +6.330 |
goodbye | +6.297 |
Neg logit contributions | |
racuse | -9.424 |
ibaba | -9.409 |
WARE | -9.402 |
buquerque | -9.086 |
Vector | -9.030 |
Token | by |
Token ID | 416 |
Feature activation | +4.39e+00 |
Pos logit contributions | |
arthed | +4.832 |
ept | +4.515 |
##### | +4.433 |
goodbye | +4.296 |
Dol | +4.272 |
Neg logit contributions | |
ibaba | -5.797 |
racuse | -5.543 |
Vector | -5.538 |
WARE | -5.432 |
buquerque | -5.335 |
Token | the |
Token ID | 262 |
Feature activation | +7.70e+00 |
Pos logit contributions | |
arthed | +2.995 |
ept | +2.828 |
##### | +2.771 |
Dol | +2.696 |
goodbye | +2.654 |
Neg logit contributions | |
ibaba | -2.892 |
WARE | -2.722 |
Vector | -2.710 |
racuse | -2.669 |
buquerque | -2.613 |
Token | Philadelphia |
Token ID | 8857 |
Feature activation | +3.88e+00 |
Pos logit contributions | |
arthed | +5.439 |
ept | +5.323 |
##### | +5.263 |
Dol | +5.146 |
goodbye | +5.049 |
Neg logit contributions | |
ibaba | -4.638 |
WARE | -4.345 |
racuse | -4.172 |
Vector | -4.140 |
buquerque | -4.105 |
Token | from |
Token ID | 422 |
Feature activation | +5.84e+00 |
Pos logit contributions | |
ibaba | +0.052 |
Vector | +0.050 |
racuse | +0.050 |
WARE | +0.049 |
buquerque | +0.048 |
Neg logit contributions | |
arthed | -0.062 |
ept | -0.058 |
##### | -0.057 |
iltr | -0.055 |
Dol | -0.054 |
Token | vigorous |
Token ID | 31543 |
Feature activation | +6.51e+00 |
Pos logit contributions | |
arthed | +4.059 |
ept | +3.788 |
##### | +3.788 |
Dol | +3.710 |
goodbye | +3.600 |
Neg logit contributions | |
ibaba | -3.704 |
racuse | -3.525 |
Vector | -3.475 |
buquerque | -3.458 |
WARE | -3.444 |
Token | environmental |
Token ID | 6142 |
Feature activation | +8.55e+00 |
Pos logit contributions | |
arthed | +4.144 |
ept | +3.896 |
##### | +3.892 |
Dol | +3.847 |
goodbye | +3.790 |
Neg logit contributions | |
ibaba | -4.423 |
racuse | -4.263 |
Vector | -4.136 |
buquerque | -4.084 |
WARE | -4.072 |
Token | regulation |
Token ID | 9001 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
arthed | +5.545 |
##### | +5.202 |
ept | +5.154 |
Dol | +5.118 |
goodbye | +5.036 |
Neg logit contributions | |
ibaba | -5.556 |
racuse | -5.489 |
Vector | -5.280 |
WARE | -5.256 |
buquerque | -5.179 |
Token | efforts |
Token ID | 4040 |
Feature activation | -5.66e-01 |
Pos logit contributions | |
arthed | +0.978 |
ept | +0.918 |
##### | +0.898 |
iltr | +0.877 |
Dol | +0.863 |
Neg logit contributions | |
ibaba | -0.864 |
Vector | -0.827 |
racuse | -0.825 |
WARE | -0.802 |
buquerque | -0.797 |
Token | over |
Token ID | 625 |
Feature activation | +1.90e+01 |
Pos logit contributions | |
ibaba | +0.349 |
Vector | +0.336 |
racuse | +0.330 |
WARE | +0.324 |
buquerque | +0.322 |
Neg logit contributions | |
arthed | -0.416 |
ept | -0.389 |
##### | -0.380 |
iltr | -0.373 |
Dol | -0.364 |
Token | the |
Token ID | 262 |
Feature activation | +1.27e+01 |
Pos logit contributions | |
arthed | +10.298 |
##### | +9.564 |
Dol | +9.538 |
goodbye | +9.418 |
ept | +9.359 |
Neg logit contributions | |
ibaba | -12.172 |
racuse | -11.532 |
WARE | -11.081 |
ABLE | -11.024 |
sidx | -10.888 |
Token | last |
Token ID | 938 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
arthed | +5.561 |
Dol | +5.462 |
##### | +5.383 |
goodbye | +5.157 |
ept | +5.042 |
Neg logit contributions | |
ibaba | -10.289 |
racuse | -10.127 |
WARE | -9.654 |
buquerque | -9.462 |
ABLE | -9.389 |
Token | decade |
Token ID | 5707 |
Feature activation | +5.02e+00 |
Pos logit contributions | |
arthed | +5.960 |
##### | +5.821 |
ept | +5.806 |
Dol | +5.769 |
goodbye | +5.705 |
Neg logit contributions | |
ibaba | -8.143 |
racuse | -8.046 |
WARE | -7.813 |
buquerque | -7.772 |
Vector | -7.421 |
Token | , |
Token ID | 11 |
Feature activation | +3.46e+00 |
Pos logit contributions | |
arthed | +3.324 |
ept | +3.186 |
##### | +3.087 |
iltr | +3.015 |
Dol | +2.986 |
Neg logit contributions | |
ibaba | -3.324 |
racuse | -3.189 |
Vector | -3.152 |
buquerque | -3.104 |
WARE | -3.092 |
Token | � |
Token ID | 447 |
Feature activation | -3.60e+00 |
Pos logit contributions | |
arthed | +2.511 |
ept | +2.373 |
##### | +2.335 |
iltr | +2.262 |
Dol | +2.252 |
Neg logit contributions | |
ibaba | -2.110 |
racuse | -2.011 |
Vector | -2.004 |
WARE | -1.956 |
buquerque | -1.940 |
Token | , |
Token ID | 11 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
arthed | +3.388 |
ept | +3.164 |
##### | +3.083 |
Dol | +3.037 |
goodbye | +2.984 |
Neg logit contributions | |
ibaba | -3.355 |
racuse | -3.216 |
buquerque | -3.146 |
WARE | -3.100 |
Vector | -3.094 |
Token | comprising |
Token ID | 27918 |
Feature activation | +8.12e+00 |
Pos logit contributions | |
arthed | +1.845 |
ept | +1.738 |
##### | +1.695 |
iltr | +1.657 |
Dol | +1.652 |
Neg logit contributions | |
ibaba | -1.459 |
racuse | -1.384 |
Vector | -1.368 |
WARE | -1.353 |
buquerque | -1.348 |
Token | " |
Token ID | 366 |
Feature activation | +1.29e+01 |
Pos logit contributions | |
arthed | +5.434 |
ept | +5.071 |
##### | +4.962 |
Dol | +4.939 |
goodbye | +4.831 |
Neg logit contributions | |
ibaba | -5.279 |
racuse | -5.052 |
WARE | -4.960 |
buquerque | -4.944 |
Vector | -4.733 |
Token | a |
Token ID | 64 |
Feature activation | +1.39e+01 |
Pos logit contributions | |
arthed | +8.925 |
ept | +8.661 |
iltr | +8.113 |
bell | +7.944 |
##### | +7.879 |
Neg logit contributions | |
racuse | -7.456 |
ibaba | -7.415 |
buquerque | -6.922 |
Vector | -6.865 |
Higgins | -6.803 |
Token | blend |
Token ID | 13516 |
Feature activation | +3.65e+00 |
Pos logit contributions | |
arthed | +8.634 |
ept | +8.306 |
##### | +8.216 |
Dol | +8.159 |
goodbye | +8.061 |
Neg logit contributions | |
ibaba | -8.227 |
racuse | -8.173 |
buquerque | -7.934 |
WARE | -7.793 |
Hamilton | -7.532 |
Token | of |
Token ID | 286 |
Feature activation | +1.89e+01 |
Pos logit contributions | |
arthed | +2.483 |
ept | +2.297 |
##### | +2.238 |
iltr | +2.208 |
Dol | +2.165 |
Neg logit contributions | |
ibaba | -2.429 |
racuse | -2.313 |
Vector | -2.310 |
WARE | -2.255 |
buquerque | -2.235 |
Token | retail |
Token ID | 6308 |
Feature activation | +9.81e+00 |
Pos logit contributions | |
arthed | +11.339 |
Dol | +10.502 |
Breach | +10.387 |
ept | +10.329 |
goodbye | +10.329 |
Neg logit contributions | |
ibaba | -11.197 |
racuse | -10.896 |
buquerque | -10.403 |
WARE | -10.361 |
sidx | -9.712 |
Token | packaged |
Token ID | 25555 |
Feature activation | +9.03e+00 |
Pos logit contributions | |
arthed | +6.389 |
ept | +5.981 |
Dol | +5.805 |
##### | +5.766 |
goodbye | +5.613 |
Neg logit contributions | |
ibaba | -6.348 |
racuse | -6.211 |
buquerque | -5.901 |
WARE | -5.797 |
Vector | -5.782 |
Token | goods |
Token ID | 7017 |
Feature activation | +6.29e-01 |
Pos logit contributions | |
arthed | +6.102 |
ept | +5.791 |
Dol | +5.719 |
##### | +5.615 |
goodbye | +5.503 |
Neg logit contributions | |
ibaba | -5.761 |
racuse | -5.571 |
buquerque | -5.292 |
WARE | -5.173 |
Vector | -5.124 |
Token | sales |
Token ID | 4200 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
arthed | +0.471 |
ept | +0.443 |
##### | +0.432 |
iltr | +0.424 |
Dol | +0.417 |
Neg logit contributions | |
ibaba | -0.379 |
Vector | -0.362 |
racuse | -0.359 |
WARE | -0.351 |
buquerque | -0.348 |
Token | , |
Token ID | 11 |
Feature activation | +9.38e+00 |
Pos logit contributions | |
arthed | +1.572 |
ept | +1.482 |
##### | +1.432 |
iltr | +1.407 |
Dol | +1.389 |
Neg logit contributions | |
ibaba | -1.335 |
racuse | -1.282 |
Vector | -1.270 |
WARE | -1.238 |
buquerque | -1.231 |
Token | put |
Token ID | 1234 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
arthed | +0.427 |
ept | +0.399 |
##### | +0.391 |
iltr | +0.383 |
Dol | +0.376 |
Neg logit contributions | |
ibaba | -0.372 |
Vector | -0.356 |
racuse | -0.354 |
WARE | -0.346 |
buquerque | -0.344 |
Token | the |
Token ID | 262 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
ibaba | +2.589 |
Vector | +2.556 |
Higgins | +2.419 |
racuse | +2.390 |
buquerque | +2.382 |
Neg logit contributions | |
arthed | -3.540 |
ept | -3.346 |
iltr | -3.230 |
##### | -3.217 |
Dol | -3.047 |
Token | Crown |
Token ID | 12223 |
Feature activation | +3.60e+00 |
Pos logit contributions | |
ibaba | +2.121 |
Vector | +2.095 |
racuse | +1.979 |
Higgins | +1.973 |
WARE | +1.964 |
Neg logit contributions | |
arthed | -2.484 |
ept | -2.308 |
##### | -2.224 |
iltr | -2.222 |
Dol | -2.108 |
Token | corporation |
Token ID | 12017 |
Feature activation | +8.70e-01 |
Pos logit contributions | |
arthed | +2.452 |
ept | +2.332 |
##### | +2.257 |
goodbye | +2.196 |
iltr | +2.185 |
Neg logit contributions | |
ibaba | -2.347 |
racuse | -2.248 |
Vector | -2.203 |
WARE | -2.198 |
buquerque | -2.179 |
Token | on |
Token ID | 319 |
Feature activation | -1.55e+01 |
Pos logit contributions | |
arthed | +0.632 |
ept | +0.592 |
##### | +0.580 |
iltr | +0.566 |
Dol | +0.558 |
Neg logit contributions | |
ibaba | -0.543 |
Vector | -0.520 |
racuse | -0.517 |
WARE | -0.505 |
buquerque | -0.501 |
Token | a |
Token ID | 257 |
Feature activation | -2.97e+01 |
Pos logit contributions | |
ibaba | +7.290 |
Vector | +6.565 |
Higgins | +6.161 |
Hamilton | +6.103 |
Startup | +5.921 |
Neg logit contributions | |
arthed | -12.466 |
ept | -12.001 |
iltr | -11.814 |
##### | -11.707 |
ⓘ | -11.023 |
Token | collision |
Token ID | 17661 |
Feature activation | -8.96e+00 |
Pos logit contributions | |
ibaba | +11.437 |
Higgins | +11.153 |
Hamilton | +10.979 |
Vector | +10.666 |
profit | +10.492 |
Neg logit contributions | |
arthed | -20.581 |
iltr | -18.518 |
ept | -17.664 |
olon | -17.466 |
##### | -17.314 |
Token | course |
Token ID | 1781 |
Feature activation | -5.60e+00 |
Pos logit contributions | |
Vector | +2.101 |
ibaba | +2.024 |
Higgins | +1.839 |
ABLE | +1.743 |
racuse | +1.721 |
Neg logit contributions | |
arthed | -9.684 |
ept | -9.372 |
##### | -9.079 |
iltr | -8.968 |
olon | -8.703 |
Token | with |
Token ID | 351 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
ibaba | +3.192 |
Vector | +3.153 |
Higgins | +2.996 |
racuse | +2.947 |
WARE | +2.916 |
Neg logit contributions | |
arthed | -4.369 |
ept | -4.034 |
##### | -3.946 |
iltr | -3.858 |
Dol | -3.651 |
Token | debt |
Token ID | 5057 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
arthed | +1.522 |
ept | +1.429 |
##### | +1.403 |
iltr | +1.363 |
Dol | +1.359 |
Neg logit contributions | |
ibaba | -1.414 |
racuse | -1.354 |
Vector | -1.336 |
WARE | -1.320 |
buquerque | -1.306 |
Token | that |
Token ID | 326 |
Feature activation | +4.81e-01 |
Pos logit contributions | |
ibaba | +1.245 |
Vector | +1.211 |
racuse | +1.168 |
WARE | +1.148 |
Higgins | +1.140 |
Neg logit contributions | |
arthed | -1.597 |
ept | -1.490 |
##### | -1.460 |
iltr | -1.434 |
Dol | -1.391 |
Token | . |
Token ID | 13 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
Vector | +5.951 |
Higgins | +5.624 |
ibaba | +5.599 |
buquerque | +5.421 |
Hamilton | +5.396 |
Neg logit contributions | |
arthed | -12.814 |
ept | -11.632 |
iltr | -11.488 |
goodbye | -11.132 |
##### | -11.086 |
Token | 8 |
Token ID | 23 |
Feature activation | -1.16e+01 |
Pos logit contributions | |
ibaba | +1.562 |
Vector | +1.520 |
racuse | +1.482 |
WARE | +1.443 |
buquerque | +1.433 |
Neg logit contributions | |
arthed | -2.151 |
ept | -2.020 |
##### | -1.976 |
iltr | -1.941 |
goodbye | -1.895 |
Token | mil |
Token ID | 1465 |
Feature activation | -3.73e+00 |
Pos logit contributions | |
ibaba | +4.985 |
Vector | +4.940 |
buquerque | +4.691 |
Higgins | +4.595 |
ABLE | +4.543 |
Neg logit contributions | |
arthed | -9.917 |
ept | -9.068 |
iltr | -8.977 |
ⓘ | -8.729 |
##### | -8.689 |
Token | and |
Token ID | 290 |
Feature activation | -9.54e+00 |
Pos logit contributions | |
ibaba | +2.288 |
Vector | +2.213 |
racuse | +2.154 |
buquerque | +2.107 |
WARE | +2.103 |
Neg logit contributions | |
arthed | -2.761 |
ept | -2.538 |
##### | -2.469 |
iltr | -2.450 |
Dol | -2.372 |
Token | a |
Token ID | 257 |
Feature activation | -1.51e+01 |
Pos logit contributions | |
Vector | +5.368 |
ibaba | +5.071 |
Higgins | +4.944 |
Hamilton | +4.931 |
racuse | +4.907 |
Neg logit contributions | |
arthed | -7.199 |
ept | -6.707 |
iltr | -6.492 |
##### | -6.321 |
ⓘ | -6.231 |
Token | 2 |
Token ID | 362 |
Feature activation | -2.26e+01 |
Pos logit contributions | |
Vector | +6.741 |
ibaba | +6.535 |
racuse | +6.124 |
ABLE | +6.105 |
Higgins | +6.060 |
Neg logit contributions | |
arthed | -12.369 |
iltr | -11.103 |
ept | -11.068 |
ⓘ | -10.633 |
##### | -10.529 |
Token | . |
Token ID | 13 |
Feature activation | -4.02e+00 |
Pos logit contributions | |
Vector | +7.282 |
buquerque | +6.432 |
ABLE | +6.375 |
Higgins | +6.355 |
Hamilton | +6.222 |
Neg logit contributions | |
arthed | -18.618 |
ept | -16.995 |
olon | -16.262 |
iltr | -16.236 |
##### | -16.100 |
Token | 4 |
Token ID | 19 |
Feature activation | -1.52e+01 |
Pos logit contributions | |
ibaba | +2.179 |
Vector | +2.126 |
racuse | +2.076 |
ABLE | +2.005 |
WARE | +2.003 |
Neg logit contributions | |
arthed | -3.164 |
ept | -2.978 |
##### | -2.930 |
iltr | -2.866 |
goodbye | -2.802 |
Token | in |
Token ID | 287 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
ibaba | +7.613 |
Vector | +7.541 |
ABLE | +7.106 |
racuse | +7.005 |
buquerque | +6.997 |
Neg logit contributions | |
arthed | -11.280 |
ept | -10.087 |
iltr | -9.841 |
##### | -9.743 |
ⓘ | -9.713 |
Token | March |
Token ID | 2805 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
ibaba | +5.685 |
Vector | +5.682 |
Hamilton | +5.476 |
racuse | +5.426 |
Higgins | +5.425 |
Neg logit contributions | |
arthed | -8.715 |
iltr | -8.177 |
ept | -7.915 |
##### | -7.865 |
ⓘ | -7.861 |
Token | 2002 |
Token ID | 6244 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
ibaba | +0.828 |
Vector | +0.802 |
racuse | +0.782 |
WARE | +0.767 |
buquerque | +0.765 |
Neg logit contributions | |
arthed | -1.028 |
ept | -0.962 |
##### | -0.937 |
iltr | -0.927 |
Dol | -0.901 |
Token | Assange |
Token ID | 21882 |
Feature activation | -6.67e+00 |
Pos logit contributions | |
ibaba | +1.694 |
Vector | +1.661 |
racuse | +1.599 |
Higgins | +1.579 |
buquerque | +1.566 |
Neg logit contributions | |
arthed | -2.292 |
ept | -2.145 |
##### | -2.092 |
iltr | -2.075 |
Dol | -1.987 |
Token | would |
Token ID | 561 |
Feature activation | -5.82e+00 |
Pos logit contributions | |
Vector | +3.308 |
ibaba | +3.287 |
Higgins | +3.154 |
Startup | +3.047 |
buquerque | +3.034 |
Neg logit contributions | |
arthed | -5.512 |
ept | -5.102 |
iltr | -4.966 |
##### | -4.939 |
ⓘ | -4.737 |
Token | be |
Token ID | 307 |
Feature activation | -5.33e-01 |
Pos logit contributions | |
Vector | +3.224 |
ibaba | +3.203 |
Higgins | +3.006 |
racuse | +3.004 |
ABLE | +2.990 |
Neg logit contributions | |
arthed | -4.576 |
ept | -4.148 |
##### | -3.981 |
iltr | -3.962 |
ⓘ | -3.858 |
Token | put |
Token ID | 1234 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
ibaba | +0.323 |
Vector | +0.311 |
racuse | +0.305 |
WARE | +0.299 |
buquerque | +0.297 |
Neg logit contributions | |
arthed | -0.398 |
ept | -0.372 |
##### | -0.364 |
iltr | -0.358 |
Dol | -0.348 |
Token | on |
Token ID | 319 |
Feature activation | -1.90e+01 |
Pos logit contributions | |
ibaba | +0.201 |
Vector | +0.193 |
racuse | +0.190 |
WARE | +0.186 |
buquerque | +0.185 |
Neg logit contributions | |
arthed | -0.245 |
ept | -0.229 |
##### | -0.224 |
iltr | -0.220 |
Dol | -0.215 |
Token | a |
Token ID | 257 |
Feature activation | -2.18e+01 |
Pos logit contributions | |
Vector | +8.417 |
ibaba | +8.410 |
Higgins | +8.038 |
Hamilton | +7.831 |
Startup | +7.759 |
Neg logit contributions | |
arthed | -13.474 |
ept | -13.143 |
iltr | -12.837 |
##### | -12.752 |
ⓘ | -12.330 |
Token | plane |
Token ID | 6614 |
Feature activation | -4.03e+00 |
Pos logit contributions | |
Vector | +10.592 |
Higgins | +10.176 |
Startup | +10.152 |
Notting | +10.025 |
Hamilton | +9.844 |
Neg logit contributions | |
arthed | -14.763 |
iltr | -13.255 |
ept | -13.233 |
ⓘ | -12.616 |
##### | -12.392 |
Token | to |
Token ID | 284 |
Feature activation | -6.10e+00 |
Pos logit contributions | |
ibaba | +2.332 |
Vector | +2.285 |
racuse | +2.197 |
Higgins | +2.167 |
ABLE | +2.149 |
Neg logit contributions | |
arthed | -3.068 |
ept | -2.857 |
##### | -2.786 |
iltr | -2.754 |
Dol | -2.642 |
Token | Sweden |
Token ID | 10710 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
Vector | +3.198 |
ibaba | +3.198 |
Higgins | +3.055 |
racuse | +3.024 |
buquerque | +3.014 |
Neg logit contributions | |
arthed | -4.920 |
ept | -4.560 |
iltr | -4.417 |
##### | -4.369 |
goodbye | -4.150 |
Token | almost |
Token ID | 2048 |
Feature activation | -3.47e+00 |
Pos logit contributions | |
arthed | +0.532 |
ept | +0.498 |
##### | +0.486 |
iltr | +0.476 |
Dol | +0.468 |
Neg logit contributions | |
ibaba | -0.471 |
Vector | -0.451 |
racuse | -0.446 |
WARE | -0.438 |
buquerque | -0.434 |
Token | straight |
Token ID | 3892 |
Feature activation | +5.61e+00 |
Pos logit contributions | |
ibaba | +1.897 |
Vector | +1.853 |
racuse | +1.774 |
Higgins | +1.748 |
WARE | +1.736 |
Neg logit contributions | |
arthed | -2.736 |
ept | -2.604 |
##### | -2.525 |
iltr | -2.500 |
Dol | -2.398 |
Token | sure |
Token ID | 1654 |
Feature activation | -1.30e+01 |
Pos logit contributions | |
Vector | +5.647 |
ibaba | +5.077 |
Startup | +4.919 |
Higgins | +4.880 |
ABLE | +4.863 |
Neg logit contributions | |
arthed | -9.653 |
ept | -8.795 |
iltr | -8.673 |
##### | -8.355 |
CTV | -8.157 |
Token | Amazon |
Token ID | 6186 |
Feature activation | -1.79e+01 |
Pos logit contributions | |
Vector | +6.356 |
ibaba | +5.916 |
WARE | +5.703 |
viruses | +5.610 |
Higgins | +5.589 |
Neg logit contributions | |
arthed | -10.496 |
ept | -9.498 |
iltr | -9.480 |
##### | -9.071 |
brate | -8.722 |
Token | doesn |
Token ID | 1595 |
Feature activation | -3.17e+00 |
Pos logit contributions | |
Vector | +9.128 |
WARE | +8.586 |
viruses | +8.432 |
ibaba | +8.297 |
Predators | +8.038 |
Neg logit contributions | |
arthed | -12.920 |
ept | -11.378 |
iltr | -11.291 |
brate | -10.759 |
##### | -10.685 |
Token | � |
Token ID | 447 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
ibaba | +1.711 |
Vector | +1.680 |
racuse | +1.615 |
WARE | +1.607 |
ABLE | +1.582 |
Neg logit contributions | |
arthed | -2.530 |
ept | -2.368 |
iltr | -2.291 |
##### | -2.270 |
Dol | -2.213 |
Token | � |
Token ID | 247 |
Feature activation | -7.16e+00 |
Pos logit contributions | |
ibaba | +0.169 |
Vector | +0.160 |
racuse | +0.157 |
WARE | +0.153 |
buquerque | +0.151 |
Neg logit contributions | |
arthed | -0.312 |
ept | -0.295 |
##### | -0.289 |
iltr | -0.285 |
Dol | -0.280 |
Token | t |
Token ID | 83 |
Feature activation | -2.07e+01 |
Pos logit contributions | |
ibaba | +5.184 |
Vector | +4.902 |
racuse | +4.835 |
buquerque | +4.722 |
Startup | +4.720 |
Neg logit contributions | |
arthed | -4.336 |
##### | -3.840 |
ept | -3.758 |
iltr | -3.744 |
Dol | -3.703 |
Token | remove |
Token ID | 4781 |
Feature activation | -1.79e+01 |
Pos logit contributions | |
Vector | +10.365 |
viruses | +9.598 |
profit | +9.363 |
unsafe | +8.944 |
WARE | +8.857 |
Neg logit contributions | |
arthed | -14.923 |
ept | -13.152 |
iltr | -12.294 |
onement | -12.122 |
brate | -12.068 |
Token | any |
Token ID | 597 |
Feature activation | -1.65e+01 |
Pos logit contributions | |
Vector | +8.917 |
viruses | +8.500 |
Higgins | +7.977 |
ABLE | +7.924 |
Pages | +7.840 |
Neg logit contributions | |
arthed | -12.886 |
ept | -12.136 |
iltr | -11.877 |
igne | -11.142 |
CTV | -11.112 |
Token | in |
Token ID | 287 |
Feature activation | -9.32e+00 |
Pos logit contributions | |
Vector | +8.562 |
viruses | +8.081 |
Higgins | +7.674 |
WARE | +7.645 |
ABLE | +7.605 |
Neg logit contributions | |
arthed | -12.008 |
ept | -11.101 |
iltr | -10.932 |
igne | -10.266 |
##### | -10.053 |
Token | the |
Token ID | 262 |
Feature activation | -4.17e+00 |
Pos logit contributions | |
Vector | +4.910 |
ibaba | +4.778 |
WARE | +4.656 |
ABLE | +4.515 |
Higgins | +4.507 |
Neg logit contributions | |
arthed | -7.188 |
iltr | -6.509 |
ept | -6.444 |
##### | -6.372 |
goodbye | -6.311 |
Token | future |
Token ID | 2003 |
Feature activation | -1.51e+01 |
Pos logit contributions | |
Vector | +2.078 |
ibaba | +2.074 |
WARE | +1.935 |
Higgins | +1.890 |
buquerque | +1.888 |
Neg logit contributions | |
arthed | -3.512 |
ept | -3.304 |
iltr | -3.212 |
##### | -3.169 |
goodbye | -3.091 |
Token | That |
Token ID | 2504 |
Feature activation | -4.44e+00 |
Pos logit contributions | |
ibaba | +3.055 |
Vector | +3.025 |
WARE | +3.016 |
ABLE | +2.990 |
racuse | +2.865 |
Neg logit contributions | |
arthed | -4.343 |
ept | -4.039 |
##### | -3.999 |
iltr | -3.957 |
Dol | -3.853 |
Token | means |
Token ID | 1724 |
Feature activation | -8.00e+00 |
Pos logit contributions | |
ibaba | +2.558 |
Vector | +2.464 |
racuse | +2.353 |
WARE | +2.331 |
buquerque | +2.308 |
Neg logit contributions | |
arthed | -3.425 |
ept | -3.159 |
iltr | -3.109 |
##### | -3.108 |
Dol | -2.956 |
Token | you |
Token ID | 345 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
Vector | +4.267 |
ibaba | +4.144 |
WARE | +3.935 |
Higgins | +3.917 |
ABLE | +3.888 |
Neg logit contributions | |
arthed | -6.341 |
ept | -5.868 |
iltr | -5.775 |
##### | -5.673 |
ⓘ | -5.392 |
Token | 'll |
Token ID | 1183 |
Feature activation | -9.95e+00 |
Pos logit contributions | |
ibaba | +1.262 |
Vector | +1.237 |
racuse | +1.196 |
WARE | +1.171 |
Higgins | +1.162 |
Neg logit contributions | |
arthed | -1.672 |
ept | -1.547 |
##### | -1.498 |
iltr | -1.487 |
Dol | -1.446 |
Token | likely |
Token ID | 1884 |
Feature activation | -8.34e+00 |
Pos logit contributions | |
Vector | +5.528 |
ibaba | +5.105 |
Higgins | +5.004 |
WARE | +4.960 |
racuse | +4.947 |
Neg logit contributions | |
arthed | -7.772 |
ept | -6.917 |
iltr | -6.687 |
##### | -6.532 |
ⓘ | -6.403 |
Token | receive |
Token ID | 3328 |
Feature activation | -2.03e+01 |
Pos logit contributions | |
Vector | +4.701 |
ibaba | +4.447 |
WARE | +4.319 |
Higgins | +4.272 |
racuse | +4.264 |
Neg logit contributions | |
arthed | -6.467 |
ept | -5.840 |
iltr | -5.645 |
##### | -5.499 |
Dol | -5.355 |
Token | the |
Token ID | 262 |
Feature activation | -1.33e+01 |
Pos logit contributions | |
Vector | +9.052 |
Startup | +8.030 |
ABLE | +7.978 |
viruses | +7.942 |
Higgins | +7.925 |
Neg logit contributions | |
arthed | -14.379 |
iltr | -13.902 |
ept | -13.484 |
igne | -13.051 |
##### | -13.012 |
Token | game |
Token ID | 983 |
Feature activation | -7.02e+00 |
Pos logit contributions | |
Vector | +5.899 |
Startup | +5.128 |
Hamilton | +5.100 |
Higgins | +5.006 |
Alone | +5.006 |
Neg logit contributions | |
arthed | -10.980 |
ept | -10.256 |
iltr | -10.152 |
##### | -10.059 |
igne | -9.859 |
Token | in |
Token ID | 287 |
Feature activation | -8.01e+00 |
Pos logit contributions | |
Vector | +3.712 |
ibaba | +3.664 |
racuse | +3.440 |
WARE | +3.406 |
ABLE | +3.377 |
Neg logit contributions | |
arthed | -5.514 |
ept | -5.085 |
iltr | -5.071 |
##### | -5.041 |
ⓘ | -4.800 |
Token | as |
Token ID | 355 |
Feature activation | -5.30e+00 |
Pos logit contributions | |
Vector | +4.273 |
ibaba | +4.154 |
racuse | +3.993 |
ABLE | +3.895 |
Hamilton | +3.890 |
Neg logit contributions | |
arthed | -6.262 |
iltr | -5.721 |
ept | -5.624 |
##### | -5.569 |
ⓘ | -5.381 |
Token | little |
Token ID | 1310 |
Feature activation | -6.09e-01 |
Pos logit contributions | |
ibaba | +2.667 |
Vector | +2.592 |
WARE | +2.451 |
buquerque | +2.431 |
racuse | +2.417 |
Neg logit contributions | |
arthed | -4.366 |
##### | -4.048 |
iltr | -4.008 |
ept | -4.008 |
goodbye | -3.861 |
Token | * |
Token ID | 1635 |
Feature activation | -1.40e+01 |
Pos logit contributions | |
ibaba | +1.847 |
Vector | +1.824 |
racuse | +1.750 |
WARE | +1.732 |
ABLE | +1.693 |
Neg logit contributions | |
arthed | -2.323 |
ept | -2.155 |
##### | -2.104 |
iltr | -2.066 |
goodbye | -2.033 |
Token | Get |
Token ID | 3497 |
Feature activation | -1.33e+01 |
Pos logit contributions | |
Vector | +7.773 |
Startup | +6.502 |
racuse | +6.424 |
Higgins | +6.357 |
Hamilton | +6.321 |
Neg logit contributions | |
arthed | -10.577 |
ept | -9.613 |
##### | -9.305 |
iltr | -9.278 |
goodbye | -9.233 |
Token | Volume |
Token ID | 31715 |
Feature activation | -8.11e+00 |
Pos logit contributions | |
Vector | +6.213 |
ABLE | +5.343 |
ibaba | +5.241 |
racuse | +5.224 |
DER | +5.047 |
Neg logit contributions | |
arthed | -10.809 |
goodbye | -10.284 |
##### | -10.143 |
ept | -10.007 |
iltr | -9.884 |
Token | () |
Token ID | 3419 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
Vector | +4.518 |
ibaba | +4.156 |
ABLE | +4.113 |
WARE | +4.021 |
racuse | +3.997 |
Neg logit contributions | |
arthed | -6.259 |
ept | -5.819 |
##### | -5.675 |
iltr | -5.620 |
goodbye | -5.599 |
Token | ; |
Token ID | 2162 |
Feature activation | -1.27e+01 |
Pos logit contributions | |
ibaba | +1.363 |
Vector | +1.338 |
racuse | +1.294 |
WARE | +1.273 |
buquerque | +1.255 |
Neg logit contributions | |
arthed | -1.561 |
ept | -1.442 |
##### | -1.405 |
iltr | -1.390 |
Dol | -1.348 |
Token | return |
Token ID | 1441 |
Feature activation | -2.03e+01 |
Pos logit contributions | |
Vector | +7.184 |
Startup | +5.886 |
Hamilton | +5.794 |
Higgins | +5.700 |
WARE | +5.575 |
Neg logit contributions | |
arthed | -10.115 |
ept | -9.129 |
iltr | -8.868 |
##### | -8.605 |
agements | -8.570 |
Token | f |
Token ID | 277 |
Feature activation | -8.71e+00 |
Pos logit contributions | |
Vector | +11.317 |
Startup | +8.934 |
Hamilton | +8.749 |
Higgins | +8.580 |
unsafe | +8.390 |
Neg logit contributions | |
arthed | -14.497 |
iltr | -12.971 |
ept | -12.855 |
onement | -12.690 |
##### | -12.575 |
Token | Mass |
Token ID | 20273 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
Vector | +2.172 |
ibaba | +1.935 |
racuse | +1.897 |
WARE | +1.680 |
buquerque | +1.640 |
Neg logit contributions | |
arthed | -9.249 |
ept | -8.792 |
##### | -8.719 |
goodbye | -8.649 |
iltr | -8.612 |
Token | ; |
Token ID | 2162 |
Feature activation | -6.10e+00 |
Pos logit contributions | |
ibaba | +1.405 |
Vector | +1.390 |
racuse | +1.335 |
WARE | +1.320 |
buquerque | +1.300 |
Neg logit contributions | |
arthed | -1.719 |
ept | -1.599 |
##### | -1.557 |
iltr | -1.540 |
Dol | -1.501 |
Token | } |
Token ID | 1782 |
Feature activation | -1.26e+01 |
Pos logit contributions | |
Vector | +3.038 |
ibaba | +2.811 |
racuse | +2.718 |
WARE | +2.695 |
Startup | +2.622 |
Neg logit contributions | |
arthed | -5.193 |
ept | -4.852 |
iltr | -4.699 |
##### | -4.698 |
Dol | -4.555 |
Token |
|
Token ID | 198 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
Vector | +7.434 |
Startup | +6.410 |
Higgins | +6.229 |
WARE | +6.216 |
Hamilton | +6.126 |
Neg logit contributions | |
arthed | -9.344 |
ept | -8.776 |
iltr | -8.409 |
onement | -8.044 |
agements | -8.029 |
Token | ; |
Token ID | 2162 |
Feature activation | -1.47e+01 |
Pos logit contributions | |
Vector | +3.093 |
ibaba | +3.005 |
racuse | +2.878 |
WARE | +2.842 |
buquerque | +2.792 |
Neg logit contributions | |
arthed | -3.828 |
ept | -3.561 |
##### | -3.436 |
iltr | -3.414 |
Dol | -3.316 |
Token | density |
Token ID | 12109 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
Vector | +8.256 |
WARE | +6.461 |
Startup | +6.380 |
Hamilton | +6.248 |
Higgins | +6.232 |
Neg logit contributions | |
arthed | -11.470 |
ept | -10.556 |
iltr | -10.167 |
##### | -9.762 |
onement | -9.616 |
Token | Grid |
Token ID | 41339 |
Feature activation | +4.93e-01 |
Pos logit contributions | |
Vector | +2.114 |
ibaba | +2.078 |
WARE | +2.008 |
racuse | +1.990 |
ABLE | +1.953 |
Neg logit contributions | |
arthed | -2.665 |
ept | -2.468 |
##### | -2.401 |
iltr | -2.397 |
Dol | -2.345 |
Token | . |
Token ID | 13 |
Feature activation | -4.06e+00 |
Pos logit contributions | |
arthed | +0.358 |
ept | +0.335 |
##### | +0.328 |
iltr | +0.321 |
Dol | +0.314 |
Neg logit contributions | |
ibaba | -0.310 |
Vector | -0.296 |
racuse | -0.293 |
WARE | -0.287 |
buquerque | -0.285 |
Token | Init |
Token ID | 31768 |
Feature activation | -6.17e+00 |
Pos logit contributions | |
Vector | +2.494 |
ibaba | +2.443 |
WARE | +2.365 |
racuse | +2.355 |
ABLE | +2.320 |
Neg logit contributions | |
arthed | -2.940 |
ept | -2.717 |
##### | -2.666 |
iltr | -2.637 |
Dol | -2.548 |
Token | ( |
Token ID | 7 |
Feature activation | -2.01e+01 |
Pos logit contributions | |
Vector | +3.802 |
ibaba | +3.583 |
racuse | +3.508 |
WARE | +3.487 |
ABLE | +3.470 |
Neg logit contributions | |
arthed | -4.452 |
ept | -4.146 |
##### | -4.059 |
iltr | -4.023 |
goodbye | -3.935 |
Token | 0 |
Token ID | 657 |
Feature activation | -6.13e+00 |
Pos logit contributions | |
Vector | +12.321 |
racuse | +9.682 |
Startup | +9.569 |
Higgins | +9.459 |
Hamilton | +9.363 |
Neg logit contributions | |
arthed | -13.082 |
ept | -11.970 |
iltr | -11.797 |
##### | -11.177 |
ⓘ | -11.085 |
Token | . |
Token ID | 13 |
Feature activation | -7.17e+00 |
Pos logit contributions | |
Vector | +3.250 |
ibaba | +3.189 |
racuse | +3.070 |
WARE | +3.055 |
buquerque | +3.018 |
Neg logit contributions | |
arthed | -4.843 |
ept | -4.498 |
iltr | -4.363 |
##### | -4.353 |
goodbye | -4.234 |
Token | 0 |
Token ID | 15 |
Feature activation | -1.77e+01 |
Pos logit contributions | |
Vector | +3.887 |
ibaba | +3.685 |
racuse | +3.627 |
ABLE | +3.622 |
WARE | +3.587 |
Neg logit contributions | |
arthed | -5.603 |
ept | -5.220 |
iltr | -5.029 |
##### | -4.970 |
goodbye | -4.836 |
Token | f |
Token ID | 69 |
Feature activation | -1.37e+01 |
Pos logit contributions | |
Vector | +7.119 |
WARE | +6.554 |
sidx | +6.521 |
Higgins | +6.434 |
Hamilton | +6.260 |
Neg logit contributions | |
arthed | -14.242 |
ept | -13.033 |
iltr | -12.597 |
##### | -12.482 |
onement | -12.440 |
Token | ) |
Token ID | 1267 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
Vector | +6.938 |
racuse | +6.208 |
WARE | +6.148 |
ibaba | +6.088 |
Higgins | +6.042 |
Neg logit contributions | |
arthed | -10.665 |
ept | -9.938 |
iltr | -9.444 |
##### | -9.295 |
goodbye | -9.157 |
Token | . |
Token ID | 13 |
Feature activation | -6.16e+00 |
Pos logit contributions | |
ibaba | +7.083 |
Vector | +6.861 |
WARE | +6.600 |
buquerque | +6.473 |
ABLE | +6.272 |
Neg logit contributions | |
arthed | -11.024 |
ept | -9.619 |
##### | -9.468 |
iltr | -9.448 |
onement | -9.247 |
Token | An |
Token ID | 1052 |
Feature activation | -5.10e+00 |
Pos logit contributions | |
Vector | +3.335 |
ibaba | +3.280 |
WARE | +3.152 |
Startup | +3.118 |
Higgins | +3.065 |
Neg logit contributions | |
arthed | -4.961 |
ept | -4.542 |
iltr | -4.412 |
##### | -4.335 |
goodbye | -4.258 |
Token | array |
Token ID | 7177 |
Feature activation | -6.80e+00 |
Pos logit contributions | |
ibaba | +2.850 |
Vector | +2.655 |
WARE | +2.555 |
racuse | +2.534 |
buquerque | +2.499 |
Neg logit contributions | |
arthed | -4.072 |
ept | -3.763 |
##### | -3.724 |
iltr | -3.631 |
Dol | -3.577 |
Token | of |
Token ID | 286 |
Feature activation | -1.00e+01 |
Pos logit contributions | |
ibaba | +3.569 |
Vector | +3.441 |
WARE | +3.222 |
racuse | +3.204 |
ABLE | +3.190 |
Neg logit contributions | |
arthed | -5.536 |
iltr | -5.073 |
ept | -5.056 |
##### | -4.947 |
goodbye | -4.769 |
Token | online |
Token ID | 2691 |
Feature activation | -1.48e+01 |
Pos logit contributions | |
Vector | +5.215 |
ibaba | +5.213 |
WARE | +4.790 |
Startup | +4.742 |
ABLE | +4.728 |
Neg logit contributions | |
arthed | -7.994 |
iltr | -7.258 |
ept | -7.189 |
##### | -6.980 |
igne | -6.762 |
Token | stores |
Token ID | 7000 |
Feature activation | -2.00e+01 |
Pos logit contributions | |
ibaba | +7.186 |
Vector | +6.824 |
WARE | +6.574 |
viruses | +6.348 |
ABLE | +6.305 |
Neg logit contributions | |
arthed | -11.772 |
iltr | -10.660 |
ept | -10.531 |
##### | -10.040 |
igne | -9.993 |
Token | offer |
Token ID | 2897 |
Feature activation | -8.85e+00 |
Pos logit contributions | |
ibaba | +8.767 |
Vector | +8.327 |
viruses | +7.629 |
Hamilton | +7.580 |
Higgins | +7.575 |
Neg logit contributions | |
arthed | -15.485 |
iltr | -12.948 |
ept | -12.832 |
##### | -12.775 |
CTV | -12.651 |
Token | a |
Token ID | 257 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
ibaba | +5.006 |
Vector | +4.969 |
WARE | +4.676 |
ABLE | +4.623 |
Higgins | +4.584 |
Neg logit contributions | |
arthed | -6.662 |
ept | -6.019 |
iltr | -6.013 |
##### | -5.927 |
igne | -5.562 |
Token | huge |
Token ID | 3236 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ibaba | +0.181 |
Vector | +0.175 |
racuse | +0.172 |
WARE | +0.169 |
buquerque | +0.168 |
Neg logit contributions | |
arthed | -0.173 |
ept | -0.161 |
##### | -0.157 |
iltr | -0.154 |
Dol | -0.150 |
Token | range |
Token ID | 2837 |
Feature activation | -7.62e+00 |
Pos logit contributions | |
arthed | +0.734 |
ept | +0.686 |
##### | +0.670 |
iltr | +0.653 |
Dol | +0.642 |
Neg logit contributions | |
ibaba | -0.769 |
Vector | -0.741 |
racuse | -0.737 |
WARE | -0.722 |
buquerque | -0.718 |
Token | of |
Token ID | 286 |
Feature activation | -7.45e+00 |
Pos logit contributions | |
ibaba | +4.049 |
Vector | +3.859 |
racuse | +3.682 |
WARE | +3.666 |
buquerque | +3.625 |
Neg logit contributions | |
arthed | -6.124 |
ept | -5.655 |
iltr | -5.496 |
##### | -5.491 |
goodbye | -5.312 |
Token | storms |
Token ID | 20228 |
Feature activation | +8.01e-01 |
Pos logit contributions | |
ibaba | +2.087 |
Vector | +2.023 |
racuse | +1.961 |
WARE | +1.926 |
buquerque | +1.923 |
Neg logit contributions | |
arthed | -2.467 |
ept | -2.267 |
##### | -2.206 |
iltr | -2.182 |
Dol | -2.126 |
Token | of |
Token ID | 286 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
arthed | +0.578 |
ept | +0.542 |
##### | +0.531 |
iltr | +0.519 |
Dol | +0.509 |
Neg logit contributions | |
ibaba | -0.503 |
Vector | -0.483 |
racuse | -0.479 |
WARE | -0.468 |
buquerque | -0.465 |
Token | revolt |
Token ID | 27360 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
arthed | +1.592 |
ept | +1.502 |
##### | +1.472 |
iltr | +1.429 |
Dol | +1.415 |
Neg logit contributions | |
ibaba | -1.392 |
Vector | -1.333 |
racuse | -1.327 |
WARE | -1.302 |
buquerque | -1.287 |
Token | than |
Token ID | 621 |
Feature activation | -7.05e+00 |
Pos logit contributions | |
arthed | +1.925 |
ept | +1.813 |
##### | +1.780 |
iltr | +1.737 |
Dol | +1.705 |
Neg logit contributions | |
ibaba | -1.786 |
racuse | -1.708 |
Vector | -1.707 |
WARE | -1.670 |
buquerque | -1.657 |
Token | the |
Token ID | 262 |
Feature activation | -5.90e+00 |
Pos logit contributions | |
ibaba | +3.980 |
Vector | +3.829 |
racuse | +3.692 |
buquerque | +3.629 |
WARE | +3.584 |
Neg logit contributions | |
arthed | -5.388 |
ept | -4.980 |
iltr | -4.880 |
##### | -4.836 |
Dol | -4.601 |
Token | un |
Token ID | 555 |
Feature activation | -1.99e+01 |
Pos logit contributions | |
ibaba | +3.672 |
Vector | +3.629 |
racuse | +3.448 |
WARE | +3.401 |
buquerque | +3.389 |
Neg logit contributions | |
arthed | -4.242 |
ept | -3.861 |
iltr | -3.772 |
##### | -3.716 |
goodbye | -3.589 |
Token | ifying |
Token ID | 4035 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
ibaba | +10.588 |
ifiable | +9.561 |
Vector | +9.510 |
buquerque | +9.498 |
life | +9.239 |
Neg logit contributions | |
arthed | -12.600 |
Dol | -11.517 |
##### | -11.418 |
goodbye | -11.191 |
ⓘ | -11.064 |
Token | tut |
Token ID | 9732 |
Feature activation | +8.80e-01 |
Pos logit contributions | |
ibaba | +1.328 |
Vector | +1.294 |
racuse | +1.246 |
WARE | +1.226 |
buquerque | +1.224 |
Neg logit contributions | |
arthed | -1.601 |
ept | -1.485 |
##### | -1.445 |
iltr | -1.434 |
Dol | -1.382 |
Token | el |
Token ID | 417 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
arthed | +0.545 |
ept | +0.508 |
##### | +0.493 |
iltr | +0.482 |
Dol | +0.469 |
Neg logit contributions | |
ibaba | -0.642 |
Vector | -0.620 |
racuse | -0.614 |
WARE | -0.605 |
buquerque | -0.600 |
Token | age |
Token ID | 496 |
Feature activation | -8.97e+00 |
Pos logit contributions | |
ibaba | +0.972 |
Vector | +0.943 |
racuse | +0.930 |
WARE | +0.916 |
buquerque | +0.912 |
Neg logit contributions | |
arthed | -0.665 |
ept | -0.605 |
##### | -0.585 |
iltr | -0.571 |
Dol | -0.555 |
Token | of |
Token ID | 286 |
Feature activation | -3.28e+00 |
Pos logit contributions | |
ibaba | +4.619 |
Vector | +4.615 |
ABLE | +4.279 |
WARE | +4.257 |
Higgins | +4.257 |
Neg logit contributions | |
arthed | -7.079 |
ept | -6.501 |
iltr | -6.304 |
##### | -6.243 |
goodbye | -6.060 |
Token | gave |
Token ID | 2921 |
Feature activation | -1.57e+01 |
Pos logit contributions | |
ibaba | +5.026 |
Vector | +4.987 |
Hamilton | +4.748 |
Higgins | +4.622 |
Notting | +4.485 |
Neg logit contributions | |
arthed | -8.593 |
ept | -8.025 |
iltr | -7.944 |
##### | -7.731 |
ⓘ | -7.445 |
Token | me |
Token ID | 502 |
Feature activation | -1.90e+01 |
Pos logit contributions | |
Vector | +6.644 |
ibaba | +6.579 |
Hamilton | +6.404 |
Higgins | +6.158 |
viruses | +6.091 |
Neg logit contributions | |
arthed | -12.845 |
iltr | -12.254 |
ept | -11.890 |
##### | -11.849 |
ⓘ | -11.317 |
Token | the |
Token ID | 262 |
Feature activation | -1.66e+01 |
Pos logit contributions | |
ibaba | +8.612 |
Hamilton | +8.405 |
Vector | +8.254 |
Higgins | +7.969 |
viruses | +7.804 |
Neg logit contributions | |
arthed | -14.141 |
iltr | -13.778 |
ept | -12.906 |
##### | -12.555 |
CTV | -12.216 |
Token | most |
Token ID | 749 |
Feature activation | -1.73e+01 |
Pos logit contributions | |
Vector | +8.075 |
ibaba | +7.893 |
Hamilton | +7.852 |
viruses | +7.787 |
Higgins | +7.723 |
Neg logit contributions | |
arthed | -12.570 |
iltr | -11.630 |
ept | -11.231 |
CTV | -10.676 |
brate | -10.512 |
Token | satisfaction |
Token ID | 14676 |
Feature activation | -1.65e+01 |
Pos logit contributions | |
ibaba | +8.595 |
Vector | +8.440 |
Hamilton | +8.077 |
viruses | +8.043 |
Higgins | +7.991 |
Neg logit contributions | |
arthed | -12.539 |
iltr | -11.654 |
ept | -11.377 |
##### | -10.698 |
ⓘ | -10.525 |
Token | were |
Token ID | 547 |
Feature activation | -1.95e+01 |
Pos logit contributions | |
Vector | +6.136 |
ibaba | +5.934 |
Hamilton | +5.631 |
viruses | +5.536 |
ABLE | +5.516 |
Neg logit contributions | |
arthed | -14.195 |
ept | -13.165 |
iltr | -12.791 |
##### | -12.552 |
ⓘ | -12.251 |
Token | scoring |
Token ID | 9689 |
Feature activation | -1.24e+01 |
Pos logit contributions | |
ibaba | +9.197 |
Hamilton | +8.877 |
Vector | +8.834 |
Shak | +8.797 |
Higgins | +8.621 |
Neg logit contributions | |
arthed | -14.118 |
iltr | -13.075 |
ept | -13.036 |
##### | -12.281 |
ⓘ | -12.030 |
Token | those |
Token ID | 883 |
Feature activation | -1.57e+01 |
Pos logit contributions | |
ibaba | +6.498 |
Vector | +6.286 |
Higgins | +6.151 |
Hamilton | +5.993 |
viruses | +5.846 |
Neg logit contributions | |
arthed | -9.401 |
iltr | -8.740 |
ept | -8.523 |
##### | -8.213 |
ⓘ | -8.174 |
Token | two |
Token ID | 734 |
Feature activation | -1.24e+01 |
Pos logit contributions | |
ibaba | +7.653 |
Vector | +7.533 |
Hamilton | +7.428 |
Higgins | +7.427 |
viruses | +7.185 |
Neg logit contributions | |
arthed | -11.773 |
iltr | -10.869 |
ept | -10.428 |
##### | -10.075 |
ⓘ | -10.052 |
Token | f |
Token ID | 277 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
ibaba | +6.668 |
Vector | +6.449 |
Higgins | +6.277 |
Hamilton | +6.220 |
viruses | +6.112 |
Neg logit contributions | |
arthed | -9.100 |
iltr | -8.453 |
ept | -8.130 |
##### | -8.021 |
ⓘ | -7.795 |
Token | ift |
Token ID | 2135 |
Feature activation | -4.27e+00 |
Pos logit contributions | |
arthed | +0.608 |
ept | +0.569 |
##### | +0.555 |
iltr | +0.544 |
Dol | +0.530 |
Neg logit contributions | |
ibaba | -0.532 |
Vector | -0.514 |
racuse | -0.509 |
WARE | -0.497 |
buquerque | -0.490 |
Token | redistribution |
Token ID | 41425 |
Feature activation | -7.12e+00 |
Pos logit contributions | |
Vector | +5.794 |
ibaba | +5.650 |
Higgins | +5.617 |
viruses | +5.605 |
Hamilton | +5.462 |
Neg logit contributions | |
arthed | -9.785 |
iltr | -9.451 |
ept | -9.402 |
##### | -9.221 |
goodbye | -8.921 |
Token | ( |
Token ID | 357 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
Vector | +3.746 |
ibaba | +3.739 |
WARE | +3.501 |
Higgins | +3.491 |
buquerque | +3.443 |
Neg logit contributions | |
arthed | -5.675 |
ept | -5.263 |
iltr | -5.218 |
##### | -5.038 |
goodbye | -4.914 |
Token | elect |
Token ID | 9509 |
Feature activation | -7.15e+00 |
Pos logit contributions | |
ibaba | +5.098 |
Vector | +5.088 |
ABLE | +4.884 |
buquerque | +4.841 |
WARE | +4.839 |
Neg logit contributions | |
arthed | -7.808 |
##### | -7.124 |
iltr | -7.091 |
ept | -7.061 |
goodbye | -6.969 |
Token | ronic |
Token ID | 4565 |
Feature activation | -9.25e+00 |
Pos logit contributions | |
ibaba | +3.179 |
Vector | +3.084 |
racuse | +3.063 |
buquerque | +2.970 |
Higgins | +2.906 |
Neg logit contributions | |
arthed | -6.129 |
ept | -5.874 |
##### | -5.685 |
iltr | -5.647 |
goodbye | -5.541 |
Token | or |
Token ID | 393 |
Feature activation | -1.19e+01 |
Pos logit contributions | |
Vector | +4.347 |
ibaba | +4.330 |
racuse | +4.109 |
WARE | +4.103 |
Hamilton | +4.041 |
Neg logit contributions | |
arthed | -7.525 |
ept | -7.200 |
iltr | -7.098 |
##### | -6.919 |
goodbye | -6.672 |
Token | otherwise |
Token ID | 4306 |
Feature activation | -1.93e+01 |
Pos logit contributions | |
Vector | +5.541 |
ibaba | +5.402 |
ABLE | +5.131 |
Higgins | +5.086 |
Hamilton | +5.085 |
Neg logit contributions | |
arthed | -9.321 |
ept | -9.103 |
iltr | -8.865 |
##### | -8.722 |
goodbye | -8.401 |
Token | , |
Token ID | 11 |
Feature activation | -1.50e+01 |
Pos logit contributions | |
Vector | +4.742 |
ibaba | +4.369 |
viruses | +4.208 |
buquerque | +4.193 |
Hamilton | +4.113 |
Neg logit contributions | |
arthed | -17.094 |
ept | -16.752 |
iltr | -16.353 |
##### | -16.009 |
goodbye | -15.654 |
Token | including |
Token ID | 1390 |
Feature activation | -1.34e+01 |
Pos logit contributions | |
Vector | +6.957 |
ibaba | +6.931 |
Higgins | +6.523 |
viruses | +6.483 |
Hamilton | +6.460 |
Neg logit contributions | |
arthed | -11.440 |
ept | -10.592 |
iltr | -10.514 |
##### | -10.384 |
goodbye | -9.964 |
Token | the |
Token ID | 262 |
Feature activation | -1.37e+01 |
Pos logit contributions | |
Vector | +7.239 |
ibaba | +7.108 |
viruses | +6.878 |
Higgins | +6.812 |
Startup | +6.599 |
Neg logit contributions | |
arthed | -9.523 |
ept | -9.221 |
iltr | -9.045 |
##### | -8.877 |
goodbye | -8.374 |
Token | world |
Token ID | 995 |
Feature activation | -1.39e+01 |
Pos logit contributions | |
Vector | +7.163 |
ibaba | +7.056 |
Higgins | +6.872 |
Startup | +6.646 |
viruses | +6.622 |
Neg logit contributions | |
arthed | -9.839 |
ept | -9.427 |
iltr | -9.261 |
##### | -8.996 |
goodbye | -8.816 |
Token | wide |
Token ID | 3094 |
Feature activation | -9.27e+00 |
Pos logit contributions | |
ibaba | +7.258 |
Vector | +7.016 |
Hamilton | +6.595 |
Higgins | +6.533 |
buquerque | +6.476 |
Neg logit contributions | |
arthed | -10.291 |
iltr | -9.653 |
ept | -9.233 |
##### | -9.093 |
onement | -8.847 |
Token | the |
Token ID | 262 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
ibaba | +1.960 |
Vector | +1.903 |
racuse | +1.846 |
buquerque | +1.834 |
Higgins | +1.810 |
Neg logit contributions | |
arthed | -2.474 |
ept | -2.306 |
iltr | -2.237 |
##### | -2.221 |
goodbye | -2.131 |
Token | sadd |
Token ID | 17766 |
Feature activation | -8.68e+00 |
Pos logit contributions | |
ibaba | +1.324 |
Vector | +1.295 |
racuse | +1.254 |
WARE | +1.238 |
buquerque | +1.237 |
Neg logit contributions | |
arthed | -1.480 |
ept | -1.369 |
iltr | -1.316 |
##### | -1.315 |
Dol | -1.262 |
Token | est |
Token ID | 395 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
ibaba | +3.208 |
Vector | +3.091 |
buquerque | +2.917 |
WARE | +2.873 |
Higgins | +2.816 |
Neg logit contributions | |
arthed | -8.207 |
ept | -7.615 |
iltr | -7.425 |
##### | -7.281 |
Dol | -7.238 |
Token | days |
Token ID | 1528 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
ibaba | +1.835 |
Vector | +1.797 |
WARE | +1.732 |
racuse | +1.728 |
buquerque | +1.727 |
Neg logit contributions | |
arthed | -1.684 |
ept | -1.530 |
iltr | -1.472 |
##### | -1.463 |
Dol | -1.403 |
Token | of |
Token ID | 286 |
Feature activation | -7.87e+00 |
Pos logit contributions | |
ibaba | +1.500 |
Vector | +1.455 |
WARE | +1.392 |
buquerque | +1.392 |
racuse | +1.389 |
Neg logit contributions | |
arthed | -1.927 |
ept | -1.794 |
##### | -1.734 |
iltr | -1.729 |
Dol | -1.668 |
Token | my |
Token ID | 616 |
Feature activation | -1.93e+01 |
Pos logit contributions | |
ibaba | +4.284 |
Vector | +4.260 |
Higgins | +4.116 |
buquerque | +4.101 |
Hamilton | +4.010 |
Neg logit contributions | |
arthed | -6.120 |
ept | -5.725 |
iltr | -5.523 |
##### | -5.352 |
ⓘ | -5.195 |
Token | life |
Token ID | 1204 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
Vector | +6.423 |
life | +6.204 |
Hamilton | +6.042 |
Higgins | +5.863 |
buquerque | +5.860 |
Neg logit contributions | |
arthed | -16.785 |
ept | -14.699 |
iltr | -14.472 |
CTV | -13.914 |
olon | -13.890 |
Token | . |
Token ID | 13 |
Feature activation | +4.97e+00 |
Pos logit contributions | |
arthed | +0.752 |
ept | +0.706 |
##### | +0.694 |
iltr | +0.678 |
goodbye | +0.666 |
Neg logit contributions | |
ibaba | -0.646 |
Vector | -0.618 |
racuse | -0.615 |
WARE | -0.601 |
buquerque | -0.592 |
Token | I |
Token ID | 314 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
arthed | +3.480 |
##### | +3.309 |
ept | +3.295 |
Dol | +3.196 |
goodbye | +3.140 |
Neg logit contributions | |
ibaba | -3.136 |
racuse | -3.017 |
Vector | -2.914 |
WARE | -2.887 |
buquerque | -2.854 |
Token | will |
Token ID | 481 |
Feature activation | -5.92e+00 |
Pos logit contributions | |
arthed | +0.014 |
ept | +0.013 |
##### | +0.013 |
iltr | +0.012 |
Dol | +0.012 |
Neg logit contributions | |
ibaba | -0.012 |
Vector | -0.011 |
racuse | -0.011 |
WARE | -0.011 |
buquerque | -0.011 |
Token | try |
Token ID | 1949 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
Vector | +3.629 |
ibaba | +3.569 |
Higgins | +3.391 |
buquerque | +3.370 |
racuse | +3.367 |
Neg logit contributions | |
arthed | -4.303 |
ept | -3.956 |
iltr | -3.772 |
##### | -3.686 |
Dol | -3.590 |
Token | suggest |
Token ID | 1950 |
Feature activation | +4.38e+00 |
Pos logit contributions | |
arthed | +1.116 |
ept | +1.052 |
##### | +1.035 |
iltr | +1.008 |
Dol | +0.988 |
Neg logit contributions | |
ibaba | -1.008 |
Vector | -0.959 |
racuse | -0.958 |
WARE | -0.939 |
buquerque | -0.936 |
Token | an |
Token ID | 281 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
arthed | +2.931 |
ept | +2.776 |
##### | +2.704 |
goodbye | +2.647 |
iltr | +2.631 |
Neg logit contributions | |
ibaba | -2.914 |
racuse | -2.796 |
Vector | -2.757 |
WARE | -2.741 |
buquerque | -2.729 |
Token | aggressive |
Token ID | 8361 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
arthed | +1.199 |
ept | +1.126 |
##### | +1.096 |
iltr | +1.081 |
Dol | +1.053 |
Neg logit contributions | |
ibaba | -1.038 |
Vector | -0.999 |
racuse | -0.995 |
WARE | -0.971 |
buquerque | -0.965 |
Token | , |
Token ID | 11 |
Feature activation | -4.47e+00 |
Pos logit contributions | |
ibaba | +1.661 |
Vector | +1.622 |
racuse | +1.555 |
WARE | +1.538 |
Higgins | +1.526 |
Neg logit contributions | |
arthed | -2.128 |
ept | -1.977 |
##### | -1.927 |
iltr | -1.894 |
Dol | -1.828 |
Token | arguably |
Token ID | 15242 |
Feature activation | -8.27e+00 |
Pos logit contributions | |
ibaba | +2.626 |
Vector | +2.581 |
racuse | +2.437 |
WARE | +2.426 |
Higgins | +2.416 |
Neg logit contributions | |
arthed | -3.357 |
ept | -3.119 |
##### | -3.032 |
iltr | -2.993 |
Dol | -2.879 |
Token | un |
Token ID | 555 |
Feature activation | -1.92e+01 |
Pos logit contributions | |
ibaba | +4.513 |
Vector | +4.494 |
Hamilton | +4.195 |
Higgins | +4.189 |
WARE | +4.188 |
Neg logit contributions | |
arthed | -6.316 |
ept | -5.804 |
iltr | -5.648 |
##### | -5.566 |
Dol | -5.283 |
Token | pat |
Token ID | 8071 |
Feature activation | -5.35e+00 |
Pos logit contributions | |
ibaba | +9.257 |
life | +9.170 |
ifiable | +9.109 |
Vector | +9.060 |
Hamilton | +9.030 |
Neg logit contributions | |
arthed | -12.285 |
##### | -11.451 |
goodbye | -11.262 |
ept | -11.173 |
Dol | -11.046 |
Token | ri |
Token ID | 380 |
Feature activation | +4.74e+00 |
Pos logit contributions | |
ibaba | +2.347 |
Vector | +2.266 |
racuse | +2.159 |
ABLE | +2.135 |
Higgins | +2.100 |
Neg logit contributions | |
arthed | -4.720 |
ept | -4.458 |
##### | -4.410 |
iltr | -4.326 |
Dol | -4.290 |
Token | otic |
Token ID | 6210 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
arthed | +2.131 |
ept | +2.012 |
##### | +1.885 |
iltr | +1.855 |
Dol | +1.755 |
Neg logit contributions | |
ibaba | -4.184 |
Vector | -4.070 |
racuse | -4.054 |
WARE | -4.006 |
Higgins | -3.962 |
Token | pattern |
Token ID | 3912 |
Feature activation | -5.66e-01 |
Pos logit contributions | |
ibaba | +0.760 |
Vector | +0.733 |
racuse | +0.715 |
WARE | +0.704 |
buquerque | +0.698 |
Neg logit contributions | |
arthed | -0.952 |
ept | -0.889 |
##### | -0.867 |
iltr | -0.852 |
Dol | -0.829 |
Token | of |
Token ID | 286 |
Feature activation | -4.11e+00 |
Pos logit contributions | |
ibaba | +0.351 |
Vector | +0.338 |
racuse | +0.332 |
WARE | +0.326 |
buquerque | +0.324 |
Neg logit contributions | |
arthed | -0.414 |
ept | -0.387 |
##### | -0.378 |
iltr | -0.371 |
Dol | -0.362 |
Token | owned |
Token ID | 11990 |
Feature activation | -4.99e+00 |
Pos logit contributions | |
ibaba | +4.500 |
Vector | +4.447 |
ABLE | +4.230 |
WARE | +4.139 |
racuse | +4.100 |
Neg logit contributions | |
arthed | -6.556 |
##### | -6.095 |
ept | -6.061 |
iltr | -5.987 |
goodbye | -5.943 |
Token | data |
Token ID | 1366 |
Feature activation | -6.35e+00 |
Pos logit contributions | |
Vector | +3.280 |
ibaba | +3.144 |
racuse | +3.023 |
Higgins | +3.017 |
WARE | +2.996 |
Neg logit contributions | |
arthed | -3.489 |
ept | -3.139 |
iltr | -3.135 |
##### | -3.053 |
Dol | -2.949 |
Token | that |
Token ID | 326 |
Feature activation | -1.05e+01 |
Pos logit contributions | |
Vector | +3.799 |
ibaba | +3.629 |
Higgins | +3.458 |
racuse | +3.456 |
WARE | +3.421 |
Neg logit contributions | |
arthed | -4.757 |
ept | -4.321 |
iltr | -4.287 |
##### | -4.228 |
Dol | -4.019 |
Token | has |
Token ID | 468 |
Feature activation | -1.23e+01 |
Pos logit contributions | |
Vector | +5.694 |
ibaba | +5.269 |
Higgins | +5.176 |
Hamilton | +5.112 |
buquerque | +5.064 |
Neg logit contributions | |
arthed | -8.215 |
iltr | -7.423 |
ept | -7.367 |
##### | -7.068 |
CTV | -6.926 |
Token | not |
Token ID | 407 |
Feature activation | -1.63e+01 |
Pos logit contributions | |
Vector | +6.243 |
ibaba | +5.718 |
Higgins | +5.539 |
Hamilton | +5.510 |
buquerque | +5.469 |
Neg logit contributions | |
arthed | -9.802 |
iltr | -8.901 |
ept | -8.734 |
##### | -8.333 |
CTV | -8.319 |
Token | yet |
Token ID | 1865 |
Feature activation | -1.92e+01 |
Pos logit contributions | |
Vector | +6.257 |
ables | +5.846 |
ibaba | +5.772 |
Hamilton | +5.749 |
buquerque | +5.742 |
Neg logit contributions | |
arthed | -13.664 |
iltr | -12.362 |
ept | -12.277 |
CTV | -12.061 |
Dol | -11.825 |
Token | been |
Token ID | 587 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
Vector | +7.142 |
Hamilton | +6.744 |
backed | +6.623 |
ibaba | +6.595 |
ifiable | +6.517 |
Neg logit contributions | |
arthed | -15.764 |
CTV | -14.034 |
iltr | -13.912 |
ept | -13.882 |
brate | -13.529 |
Token | published |
Token ID | 3199 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
Vector | +5.906 |
ibaba | +5.624 |
Hamilton | +5.372 |
Startup | +5.321 |
Higgins | +5.292 |
Neg logit contributions | |
arthed | -7.235 |
iltr | -6.574 |
ept | -6.531 |
brate | -6.193 |
CTV | -6.190 |
Token | in |
Token ID | 287 |
Feature activation | +5.11e-01 |
Pos logit contributions | |
ibaba | +1.533 |
Vector | +1.511 |
racuse | +1.447 |
WARE | +1.417 |
buquerque | +1.415 |
Neg logit contributions | |
arthed | -1.977 |
ept | -1.841 |
##### | -1.799 |
iltr | -1.787 |
Dol | -1.727 |
Token | a |
Token ID | 257 |
Feature activation | -7.33e+00 |
Pos logit contributions | |
arthed | +0.371 |
ept | +0.349 |
##### | +0.340 |
iltr | +0.332 |
Dol | +0.326 |
Neg logit contributions | |
ibaba | -0.319 |
Vector | -0.305 |
racuse | -0.303 |
WARE | -0.297 |
buquerque | -0.294 |
Token | concurrent |
Token ID | 24580 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
Vector | +4.214 |
ibaba | +3.992 |
Startup | +3.790 |
WARE | +3.779 |
Higgins | +3.763 |
Neg logit contributions | |
arthed | -5.590 |
iltr | -5.170 |
ept | -5.118 |
##### | -4.947 |
Dol | -4.792 |
Token | is |
Token ID | 318 |
Feature activation | +5.56e+00 |
Pos logit contributions | |
ibaba | +0.937 |
Vector | +0.906 |
racuse | +0.883 |
WARE | +0.867 |
buquerque | +0.867 |
Neg logit contributions | |
arthed | -1.161 |
ept | -1.082 |
##### | -1.056 |
iltr | -1.041 |
Dol | -1.014 |
Token | where |
Token ID | 810 |
Feature activation | -4.18e+00 |
Pos logit contributions | |
arthed | +3.465 |
ept | +3.237 |
##### | +3.219 |
goodbye | +3.212 |
Dol | +3.126 |
Neg logit contributions | |
ibaba | -3.871 |
racuse | -3.870 |
WARE | -3.680 |
buquerque | -3.648 |
Vector | -3.638 |
Token | Dallas |
Token ID | 8533 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
Vector | +2.555 |
ibaba | +2.538 |
racuse | +2.439 |
buquerque | +2.410 |
Higgins | +2.398 |
Neg logit contributions | |
arthed | -3.049 |
ept | -2.812 |
iltr | -2.744 |
##### | -2.724 |
Dol | -2.575 |
Token | ' |
Token ID | 6 |
Feature activation | -3.97e+00 |
Pos logit contributions | |
ibaba | +0.714 |
Vector | +0.690 |
racuse | +0.671 |
WARE | +0.660 |
buquerque | +0.656 |
Neg logit contributions | |
arthed | -0.947 |
ept | -0.883 |
##### | -0.862 |
iltr | -0.850 |
Dol | -0.827 |
Token | decisions |
Token ID | 5370 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
ibaba | +2.384 |
Vector | +2.378 |
racuse | +2.258 |
Higgins | +2.252 |
Startup | +2.235 |
Neg logit contributions | |
arthed | -2.922 |
ept | -2.712 |
iltr | -2.609 |
##### | -2.599 |
Dol | -2.480 |
Token | get |
Token ID | 651 |
Feature activation | -1.92e+01 |
Pos logit contributions | |
ibaba | +0.666 |
Vector | +0.645 |
racuse | +0.631 |
WARE | +0.618 |
buquerque | +0.616 |
Neg logit contributions | |
arthed | -0.780 |
ept | -0.724 |
##### | -0.706 |
iltr | -0.696 |
Dol | -0.677 |
Token | interesting |
Token ID | 3499 |
Feature activation | -1.11e+01 |
Pos logit contributions | |
Vector | +8.413 |
Higgins | +8.147 |
Hamilton | +8.016 |
Startup | +7.854 |
Mercer | +7.829 |
Neg logit contributions | |
arthed | -13.697 |
iltr | -13.447 |
ept | -13.446 |
##### | -12.633 |
brate | -12.177 |
Token | . |
Token ID | 13 |
Feature activation | +9.37e-01 |
Pos logit contributions | |
ibaba | +5.141 |
Vector | +5.110 |
ABLE | +4.651 |
Startup | +4.616 |
buquerque | +4.614 |
Neg logit contributions | |
arthed | -9.285 |
ept | -8.466 |
iltr | -8.157 |
##### | -8.119 |
Dol | -7.824 |
Token | If |
Token ID | 1002 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
arthed | +0.687 |
ept | +0.645 |
##### | +0.636 |
iltr | +0.618 |
Dol | +0.612 |
Neg logit contributions | |
ibaba | -0.579 |
racuse | -0.548 |
Vector | -0.547 |
WARE | -0.536 |
buquerque | -0.530 |
Token | Mont |
Token ID | 5575 |
Feature activation | +3.84e+00 |
Pos logit contributions | |
ibaba | +1.838 |
Vector | +1.835 |
racuse | +1.755 |
Higgins | +1.727 |
buquerque | +1.710 |
Neg logit contributions | |
arthed | -2.623 |
ept | -2.453 |
iltr | -2.382 |
##### | -2.381 |
Dol | -2.256 |
Token | a |
Token ID | 64 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
arthed | +2.450 |
ept | +2.318 |
##### | +2.237 |
iltr | +2.188 |
bell | +2.149 |
Neg logit contributions | |
ibaba | -2.690 |
Vector | -2.626 |
WARE | -2.602 |
racuse | -2.583 |
buquerque | -2.510 |
Token | primarily |
Token ID | 7525 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
arthed | +1.961 |
ept | +1.829 |
##### | +1.806 |
iltr | +1.736 |
Dol | +1.726 |
Neg logit contributions | |
ibaba | -2.013 |
Vector | -1.912 |
racuse | -1.903 |
WARE | -1.875 |
buquerque | -1.842 |
Token | at |
Token ID | 379 |
Feature activation | -6.37e+00 |
Pos logit contributions | |
arthed | +0.733 |
ept | +0.681 |
##### | +0.668 |
iltr | +0.652 |
Dol | +0.639 |
Neg logit contributions | |
ibaba | -0.738 |
Vector | -0.711 |
racuse | -0.701 |
WARE | -0.689 |
buquerque | -0.681 |
Token | those |
Token ID | 883 |
Feature activation | -1.14e+01 |
Pos logit contributions | |
Vector | +3.819 |
ibaba | +3.690 |
Higgins | +3.617 |
racuse | +3.549 |
buquerque | +3.530 |
Neg logit contributions | |
arthed | -4.671 |
ept | -4.338 |
iltr | -4.224 |
##### | -4.127 |
goodbye | -3.947 |
Token | viewers |
Token ID | 10209 |
Feature activation | -5.93e+00 |
Pos logit contributions | |
Vector | +5.997 |
Higgins | +5.633 |
Hamilton | +5.565 |
viruses | +5.444 |
WARE | +5.393 |
Neg logit contributions | |
arthed | -8.796 |
ept | -7.915 |
iltr | -7.779 |
##### | -7.597 |
ⓘ | -7.478 |
Token | who |
Token ID | 508 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
Vector | +3.277 |
ibaba | +3.185 |
Higgins | +3.112 |
WARE | +3.091 |
racuse | +3.043 |
Neg logit contributions | |
arthed | -4.617 |
ept | -4.266 |
iltr | -4.146 |
##### | -4.105 |
goodbye | -3.931 |
Token | took |
Token ID | 1718 |
Feature activation | -1.92e+01 |
Pos logit contributions | |
arthed | +1.028 |
ept | +0.968 |
##### | +0.954 |
iltr | +0.929 |
Dol | +0.913 |
Neg logit contributions | |
ibaba | -0.929 |
Vector | -0.882 |
racuse | -0.878 |
WARE | -0.862 |
buquerque | -0.854 |
Token | this |
Token ID | 428 |
Feature activation | -1.35e+01 |
Pos logit contributions | |
Vector | +9.341 |
Higgins | +8.983 |
Hamilton | +8.920 |
viruses | +8.688 |
Startup | +8.362 |
Neg logit contributions | |
arthed | -14.123 |
ept | -12.973 |
iltr | -12.180 |
##### | -11.786 |
igne | -11.494 |
Token | dated |
Token ID | 14567 |
Feature activation | -5.53e+00 |
Pos logit contributions | |
Vector | +7.194 |
Hamilton | +6.861 |
Higgins | +6.811 |
ibaba | +6.759 |
Startup | +6.499 |
Neg logit contributions | |
arthed | -10.144 |
ept | -9.262 |
iltr | -8.967 |
##### | -8.913 |
ⓘ | -8.483 |
Token | material |
Token ID | 2587 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
Vector | +3.252 |
ibaba | +3.229 |
Higgins | +3.069 |
WARE | +3.034 |
Startup | +3.019 |
Neg logit contributions | |
arthed | -4.128 |
ept | -3.813 |
iltr | -3.691 |
##### | -3.676 |
goodbye | -3.451 |
Token | as |
Token ID | 355 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
ibaba | +2.958 |
Vector | +2.913 |
Higgins | +2.786 |
racuse | +2.752 |
WARE | +2.745 |
Neg logit contributions | |
arthed | -4.245 |
ept | -3.961 |
##### | -3.823 |
iltr | -3.789 |
Dol | -3.594 |
Token | Gospel |
Token ID | 23244 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
ibaba | +1.588 |
Vector | +1.549 |
WARE | +1.478 |
racuse | +1.477 |
Higgins | +1.463 |
Neg logit contributions | |
arthed | -1.901 |
ept | -1.775 |
##### | -1.726 |
iltr | -1.705 |
Dol | -1.637 |
Token | people |
Token ID | 661 |
Feature activation | -8.48e+00 |
Pos logit contributions | |
Vector | +3.863 |
ibaba | +3.741 |
Hamilton | +3.708 |
Startup | +3.656 |
Higgins | +3.635 |
Neg logit contributions | |
arthed | -5.268 |
ept | -4.896 |
iltr | -4.724 |
##### | -4.677 |
Dol | -4.439 |
Token | who |
Token ID | 508 |
Feature activation | -7.56e-01 |
Pos logit contributions | |
ibaba | +4.258 |
Vector | +4.234 |
Hamilton | +4.082 |
racuse | +4.056 |
Higgins | +4.044 |
Neg logit contributions | |
arthed | -6.885 |
ept | -6.403 |
##### | -6.127 |
iltr | -6.076 |
Dol | -5.800 |
Token | work |
Token ID | 670 |
Feature activation | -5.72e+00 |
Pos logit contributions | |
ibaba | +0.453 |
Vector | +0.437 |
racuse | +0.429 |
WARE | +0.419 |
buquerque | +0.418 |
Neg logit contributions | |
arthed | -0.569 |
ept | -0.531 |
##### | -0.518 |
iltr | -0.510 |
Dol | -0.498 |
Token | downtown |
Token ID | 9436 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
ibaba | +3.144 |
Vector | +3.109 |
racuse | +2.947 |
buquerque | +2.936 |
Higgins | +2.910 |
Neg logit contributions | |
arthed | -4.465 |
ept | -4.128 |
##### | -3.998 |
iltr | -3.961 |
ⓘ | -3.830 |
Token | to |
Token ID | 284 |
Feature activation | -5.49e+00 |
Pos logit contributions | |
ibaba | +1.191 |
Vector | +1.153 |
racuse | +1.118 |
buquerque | +1.102 |
WARE | +1.100 |
Neg logit contributions | |
arthed | -1.475 |
ept | -1.373 |
##### | -1.339 |
iltr | -1.313 |
Dol | -1.280 |
Token | get |
Token ID | 651 |
Feature activation | -1.92e+01 |
Pos logit contributions | |
Vector | +3.179 |
ibaba | +3.159 |
Startup | +2.967 |
ABLE | +2.914 |
racuse | +2.910 |
Neg logit contributions | |
arthed | -4.242 |
ept | -3.884 |
iltr | -3.672 |
##### | -3.666 |
Dol | -3.560 |
Token | more |
Token ID | 517 |
Feature activation | -1.25e+01 |
Pos logit contributions | |
Hamilton | +8.037 |
Vector | +7.994 |
Startup | +7.685 |
viruses | +7.556 |
McKin | +7.453 |
Neg logit contributions | |
arthed | -14.051 |
ept | -13.868 |
iltr | -13.266 |
onement | -13.168 |
##### | -12.680 |
Token | active |
Token ID | 4075 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
Vector | +5.966 |
Hamilton | +5.586 |
Startup | +5.446 |
viruses | +5.421 |
McKin | +5.397 |
Neg logit contributions | |
arthed | -9.653 |
ept | -9.448 |
iltr | -9.024 |
onement | -8.858 |
##### | -8.811 |
Token | and |
Token ID | 290 |
Feature activation | -1.06e+01 |
Pos logit contributions | |
ibaba | +0.726 |
Vector | +0.701 |
racuse | +0.686 |
WARE | +0.672 |
buquerque | +0.668 |
Neg logit contributions | |
arthed | -0.922 |
ept | -0.863 |
##### | -0.843 |
iltr | -0.826 |
Dol | -0.807 |
Token | healthy |
Token ID | 5448 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
Vector | +5.647 |
Hamilton | +5.388 |
Startup | +5.373 |
ibaba | +5.370 |
buquerque | +5.266 |
Neg logit contributions | |
arthed | -8.112 |
ept | -7.521 |
iltr | -7.141 |
##### | -7.002 |
onement | -6.783 |
Token | . |
Token ID | 13 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
ibaba | +2.461 |
Vector | +2.428 |
racuse | +2.319 |
buquerque | +2.280 |
ABLE | +2.280 |
Neg logit contributions | |
arthed | -3.553 |
ept | -3.332 |
##### | -3.241 |
iltr | -3.181 |
Dol | -3.101 |
Token | � |
Token ID | 251 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
ibaba | +0.126 |
Vector | +0.120 |
racuse | +0.117 |
WARE | +0.114 |
buquerque | +0.113 |
Neg logit contributions | |
arthed | -0.226 |
ept | -0.214 |
##### | -0.210 |
iltr | -0.207 |
Dol | -0.203 |
Token | It |
Token ID | 632 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
ibaba | +0.869 |
Vector | +0.846 |
racuse | +0.818 |
WARE | +0.809 |
buquerque | +0.801 |
Neg logit contributions | |
arthed | -1.147 |
ept | -1.071 |
##### | -1.041 |
iltr | -1.035 |
goodbye | -0.998 |
Token | would |
Token ID | 561 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
arthed | +0.767 |
ept | +0.723 |
##### | +0.709 |
iltr | +0.693 |
Dol | +0.679 |
Neg logit contributions | |
ibaba | -0.638 |
Vector | -0.612 |
racuse | -0.607 |
WARE | -0.594 |
buquerque | -0.588 |
Token | make |
Token ID | 787 |
Feature activation | -7.12e+00 |
Pos logit contributions | |
arthed | +1.006 |
ept | +0.943 |
##### | +0.925 |
iltr | +0.899 |
Dol | +0.880 |
Neg logit contributions | |
ibaba | -1.140 |
Vector | -1.094 |
racuse | -1.092 |
WARE | -1.068 |
buquerque | -1.062 |
Token | Supreme |
Token ID | 5617 |
Feature activation | -3.99e+00 |
Pos logit contributions | |
Vector | +3.793 |
ibaba | +3.781 |
Higgins | +3.669 |
racuse | +3.564 |
Startup | +3.507 |
Neg logit contributions | |
arthed | -5.674 |
ept | -5.185 |
iltr | -5.089 |
##### | -5.008 |
ⓘ | -4.802 |
Token | un |
Token ID | 555 |
Feature activation | -1.90e+01 |
Pos logit contributions | |
ibaba | +2.367 |
Vector | +2.322 |
racuse | +2.188 |
WARE | +2.187 |
buquerque | +2.185 |
Neg logit contributions | |
arthed | -3.037 |
ept | -2.804 |
##### | -2.720 |
iltr | -2.717 |
Dol | -2.597 |
Token | app |
Token ID | 1324 |
Feature activation | +7.48e-01 |
Pos logit contributions | |
ibaba | +9.522 |
ABLE | +9.319 |
Vector | +9.308 |
Higgins | +9.019 |
buquerque | +8.973 |
Neg logit contributions | |
arthed | -12.534 |
##### | -11.329 |
ept | -11.314 |
iltr | -11.219 |
Dol | -11.101 |
Token | ealing |
Token ID | 26919 |
Feature activation | -6.18e+00 |
Pos logit contributions | |
arthed | +0.374 |
ept | +0.342 |
##### | +0.327 |
iltr | +0.318 |
Dol | +0.307 |
Neg logit contributions | |
ibaba | -0.637 |
Vector | -0.620 |
racuse | -0.613 |
WARE | -0.604 |
buquerque | -0.600 |
Token | , |
Token ID | 11 |
Feature activation | -7.79e+00 |
Pos logit contributions | |
ibaba | +3.372 |
Vector | +3.304 |
ABLE | +3.119 |
WARE | +3.096 |
buquerque | +3.080 |
Neg logit contributions | |
arthed | -4.907 |
ept | -4.503 |
##### | -4.384 |
iltr | -4.383 |
Dol | -4.214 |
Token | in |
Token ID | 287 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
Vector | +4.258 |
ibaba | +4.202 |
Higgins | +4.004 |
buquerque | +3.875 |
WARE | +3.860 |
Neg logit contributions | |
arthed | -6.167 |
ept | -5.634 |
iltr | -5.457 |
##### | -5.387 |
ⓘ | -5.103 |
Token | a |
Token ID | 257 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
ibaba | +0.786 |
Vector | +0.760 |
racuse | +0.742 |
WARE | +0.730 |
buquerque | +0.726 |
Neg logit contributions | |
arthed | -0.915 |
ept | -0.846 |
##### | -0.829 |
iltr | -0.819 |
Dol | -0.792 |
Token | one |
Token ID | 530 |
Feature activation | -4.46e+00 |
Pos logit contributions | |
ibaba | +2.415 |
Vector | +2.399 |
racuse | +2.274 |
Higgins | +2.261 |
Startup | +2.248 |
Neg logit contributions | |
arthed | -3.003 |
ept | -2.790 |
iltr | -2.698 |
##### | -2.685 |
goodbye | -2.585 |
Token | of |
Token ID | 286 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
ibaba | +2.635 |
Vector | +2.568 |
racuse | +2.434 |
Higgins | +2.407 |
Startup | +2.394 |
Neg logit contributions | |
arthed | -3.374 |
ept | -3.109 |
##### | -3.016 |
iltr | -2.998 |
goodbye | -2.909 |
Token | the |
Token ID | 262 |
Feature activation | -8.80e+00 |
Pos logit contributions | |
ibaba | +2.101 |
Vector | +2.038 |
racuse | +1.973 |
buquerque | +1.945 |
WARE | +1.930 |
Neg logit contributions | |
arthed | -2.648 |
ept | -2.475 |
##### | -2.390 |
iltr | -2.382 |
goodbye | -2.299 |
Token | top |
Token ID | 1353 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
ibaba | +4.882 |
Vector | +4.870 |
Startup | +4.717 |
Higgins | +4.579 |
WARE | +4.509 |
Neg logit contributions | |
arthed | -6.640 |
ept | -5.998 |
iltr | -5.748 |
CTV | -5.689 |
##### | -5.662 |
Token | selling |
Token ID | 6301 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
ibaba | +6.157 |
Vector | +6.156 |
Startup | +5.875 |
Higgins | +5.720 |
Notting | +5.706 |
Neg logit contributions | |
arthed | -8.176 |
ept | -7.355 |
iltr | -7.158 |
CTV | -6.997 |
brate | -6.944 |
Token | Kindle |
Token ID | 27114 |
Feature activation | -1.90e+01 |
Pos logit contributions | |
Vector | +6.320 |
ibaba | +6.041 |
Startup | +5.939 |
Notting | +5.800 |
Higgins | +5.797 |
Neg logit contributions | |
arthed | -10.368 |
iltr | -9.100 |
ept | -9.097 |
CTV | -8.897 |
ⓘ | -8.887 |
Token | Gold |
Token ID | 3561 |
Feature activation | -1.43e+01 |
Pos logit contributions | |
Vector | +9.357 |
viruses | +8.824 |
ibaba | +8.785 |
WARE | +8.680 |
Startup | +8.657 |
Neg logit contributions | |
arthed | -12.950 |
ept | -11.731 |
iltr | -11.603 |
dayName | -11.546 |
brate | -11.388 |
Token | Rus |
Token ID | 9223 |
Feature activation | -5.54e-01 |
Pos logit contributions | |
Vector | +6.324 |
ibaba | +6.292 |
racuse | +5.954 |
Startup | +5.902 |
viruses | +5.878 |
Neg logit contributions | |
arthed | -11.379 |
ept | -10.406 |
goodbye | -9.890 |
iltr | -9.852 |
##### | -9.837 |
Token | hers |
Token ID | 7084 |
Feature activation | -4.47e+00 |
Pos logit contributions | |
ibaba | +0.281 |
Vector | +0.268 |
racuse | +0.263 |
WARE | +0.256 |
buquerque | +0.254 |
Neg logit contributions | |
arthed | -0.467 |
ept | -0.440 |
##### | -0.432 |
iltr | -0.425 |
Dol | -0.417 |
Token | . |
Token ID | 13 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
ibaba | +2.527 |
Vector | +2.476 |
WARE | +2.344 |
racuse | +2.341 |
buquerque | +2.311 |
Neg logit contributions | |
arthed | -3.479 |
ept | -3.223 |
iltr | -3.129 |
##### | -3.117 |
Dol | -3.001 |
Token | He |
Token ID | 679 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
ibaba | +1.732 |
Vector | +1.721 |
racuse | +1.622 |
Higgins | +1.620 |
WARE | +1.620 |
Neg logit contributions | |
arthed | -2.366 |
ept | -2.218 |
iltr | -2.136 |
##### | -2.121 |
goodbye | -2.066 |
Token | The |
Token ID | 464 |
Feature activation | -1.23e+01 |
Pos logit contributions | |
ibaba | +3.360 |
WARE | +3.339 |
Vector | +3.321 |
ABLE | +3.257 |
Hamilton | +3.250 |
Neg logit contributions | |
arthed | -5.909 |
ept | -5.530 |
iltr | -5.459 |
##### | -5.405 |
goodbye | -5.162 |
Token | two |
Token ID | 734 |
Feature activation | -1.19e+01 |
Pos logit contributions | |
ibaba | +6.765 |
Vector | +6.714 |
Higgins | +6.589 |
Hamilton | +6.576 |
viruses | +6.402 |
Neg logit contributions | |
arthed | -9.017 |
iltr | -8.140 |
ept | -8.117 |
##### | -7.890 |
brate | -7.569 |
Token | innings |
Token ID | 16267 |
Feature activation | -9.54e+00 |
Pos logit contributions | |
ibaba | +6.590 |
Vector | +6.529 |
Hamilton | +6.404 |
Higgins | +6.354 |
viruses | +6.334 |
Neg logit contributions | |
arthed | -8.508 |
iltr | -7.844 |
ept | -7.598 |
##### | -7.408 |
brate | -7.205 |
Token | that |
Token ID | 326 |
Feature activation | -1.05e+01 |
Pos logit contributions | |
ibaba | +4.946 |
Vector | +4.785 |
Higgins | +4.588 |
Hamilton | +4.557 |
Startup | +4.423 |
Neg logit contributions | |
arthed | -7.569 |
ept | -7.058 |
iltr | -6.993 |
##### | -6.882 |
ⓘ | -6.570 |
Token | gave |
Token ID | 2921 |
Feature activation | -1.57e+01 |
Pos logit contributions | |
ibaba | +5.026 |
Vector | +4.987 |
Hamilton | +4.748 |
Higgins | +4.622 |
Notting | +4.485 |
Neg logit contributions | |
arthed | -8.593 |
ept | -8.025 |
iltr | -7.944 |
##### | -7.731 |
ⓘ | -7.445 |
Token | me |
Token ID | 502 |
Feature activation | -1.90e+01 |
Pos logit contributions | |
Vector | +6.644 |
ibaba | +6.579 |
Hamilton | +6.404 |
Higgins | +6.158 |
viruses | +6.091 |
Neg logit contributions | |
arthed | -12.845 |
iltr | -12.254 |
ept | -11.890 |
##### | -11.849 |
ⓘ | -11.317 |
Token | the |
Token ID | 262 |
Feature activation | -1.66e+01 |
Pos logit contributions | |
ibaba | +8.612 |
Hamilton | +8.405 |
Vector | +8.254 |
Higgins | +7.969 |
viruses | +7.804 |
Neg logit contributions | |
arthed | -14.141 |
iltr | -13.778 |
ept | -12.906 |
##### | -12.555 |
CTV | -12.216 |
Token | most |
Token ID | 749 |
Feature activation | -1.73e+01 |
Pos logit contributions | |
Vector | +8.075 |
ibaba | +7.893 |
Hamilton | +7.852 |
viruses | +7.787 |
Higgins | +7.723 |
Neg logit contributions | |
arthed | -12.570 |
iltr | -11.630 |
ept | -11.231 |
CTV | -10.676 |
brate | -10.512 |
Token | satisfaction |
Token ID | 14676 |
Feature activation | -1.65e+01 |
Pos logit contributions | |
ibaba | +8.595 |
Vector | +8.440 |
Hamilton | +8.077 |
viruses | +8.043 |
Higgins | +7.991 |
Neg logit contributions | |
arthed | -12.539 |
iltr | -11.654 |
ept | -11.377 |
##### | -10.698 |
ⓘ | -10.525 |
Token | were |
Token ID | 547 |
Feature activation | -1.95e+01 |
Pos logit contributions | |
Vector | +6.136 |
ibaba | +5.934 |
Hamilton | +5.631 |
viruses | +5.536 |
ABLE | +5.516 |
Neg logit contributions | |
arthed | -14.195 |
ept | -13.165 |
iltr | -12.791 |
##### | -12.552 |
ⓘ | -12.251 |
Token | scoring |
Token ID | 9689 |
Feature activation | -1.24e+01 |
Pos logit contributions | |
ibaba | +9.197 |
Hamilton | +8.877 |
Vector | +8.834 |
Shak | +8.797 |
Higgins | +8.621 |
Neg logit contributions | |
arthed | -14.118 |
iltr | -13.075 |
ept | -13.036 |
##### | -12.281 |
ⓘ | -12.030 |
Token | Progressive |
Token ID | 25852 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
arthed | +3.651 |
##### | +3.335 |
Dol | +3.307 |
ept | +3.243 |
goodbye | +3.175 |
Neg logit contributions | |
ibaba | -4.892 |
racuse | -4.739 |
WARE | -4.558 |
buquerque | -4.552 |
ABLE | -4.519 |
Token | Journalism |
Token ID | 37406 |
Feature activation | -7.15e-02 |
Pos logit contributions | |
arthed | +1.029 |
ept | +0.941 |
##### | +0.921 |
Dol | +0.885 |
iltr | +0.883 |
Neg logit contributions | |
ibaba | -1.205 |
Vector | -1.145 |
racuse | -1.145 |
buquerque | -1.124 |
WARE | -1.122 |
Token | The |
Token ID | 383 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
ibaba | +0.043 |
Vector | +0.041 |
racuse | +0.040 |
WARE | +0.040 |
buquerque | +0.039 |
Neg logit contributions | |
arthed | -0.054 |
ept | -0.050 |
##### | -0.049 |
iltr | -0.048 |
Dol | -0.048 |
Token | Nation |
Token ID | 8741 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
arthed | +1.645 |
ept | +1.557 |
##### | +1.525 |
Dol | +1.477 |
iltr | +1.474 |
Neg logit contributions | |
ibaba | -1.300 |
racuse | -1.222 |
Vector | -1.220 |
WARE | -1.190 |
buquerque | -1.184 |
Token | is |
Token ID | 318 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
ibaba | +1.506 |
Vector | +1.465 |
racuse | +1.426 |
Higgins | +1.369 |
WARE | +1.363 |
Neg logit contributions | |
arthed | -2.329 |
ept | -2.205 |
##### | -2.149 |
iltr | -2.109 |
goodbye | -2.060 |
Token | reader |
Token ID | 9173 |
Feature activation | +1.27e+01 |
Pos logit contributions | |
arthed | +1.561 |
ept | +1.438 |
##### | +1.433 |
iltr | +1.389 |
goodbye | +1.386 |
Neg logit contributions | |
ibaba | -1.274 |
racuse | -1.209 |
Vector | -1.202 |
WARE | -1.179 |
buquerque | -1.167 |
Token | supported |
Token ID | 4855 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
arthed | +7.141 |
ept | +6.619 |
##### | +6.596 |
Dol | +6.595 |
/# | +6.378 |
Neg logit contributions | |
racuse | -9.285 |
ibaba | -9.231 |
ccording | -9.072 |
WARE | -8.956 |
buquerque | -8.875 |
Token | : |
Token ID | 25 |
Feature activation | +3.37e+00 |
Pos logit contributions | |
arthed | +1.681 |
ept | +1.583 |
##### | +1.547 |
iltr | +1.504 |
Dol | +1.501 |
Neg logit contributions | |
ibaba | -1.698 |
racuse | -1.610 |
Vector | -1.603 |
WARE | -1.597 |
buquerque | -1.558 |
Token | Chip |
Token ID | 17869 |
Feature activation | +4.72e+00 |
Pos logit contributions | |
arthed | +2.372 |
##### | +2.228 |
ept | +2.185 |
goodbye | +2.112 |
Dol | +2.105 |
Neg logit contributions | |
ibaba | -2.186 |
racuse | -2.061 |
Vector | -2.020 |
WARE | -1.986 |
buquerque | -1.966 |
Token | in |
Token ID | 287 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
arthed | +2.975 |
ept | +2.797 |
##### | +2.745 |
Dol | +2.738 |
iltr | +2.709 |
Neg logit contributions | |
ibaba | -3.317 |
Vector | -3.132 |
racuse | -3.114 |
WARE | -3.093 |
buquerque | -3.034 |
Token | $ |
Token ID | 720 |
Feature activation | +8.75e-01 |
Pos logit contributions | |
ibaba | +0.034 |
Vector | +0.033 |
racuse | +0.032 |
WARE | +0.032 |
buquerque | +0.032 |
Neg logit contributions | |
arthed | -0.028 |
ept | -0.026 |
##### | -0.025 |
iltr | -0.025 |
Dol | -0.024 |
Token | that |
Token ID | 326 |
Feature activation | +5.90e+00 |
Pos logit contributions | |
arthed | +3.190 |
ept | +3.027 |
##### | +2.930 |
iltr | +2.882 |
goodbye | +2.868 |
Neg logit contributions | |
ibaba | -3.403 |
racuse | -3.321 |
Vector | -3.282 |
buquerque | -3.252 |
WARE | -3.246 |
Token | the |
Token ID | 262 |
Feature activation | +4.13e+00 |
Pos logit contributions | |
arthed | +3.884 |
ept | +3.752 |
Dol | +3.652 |
##### | +3.610 |
goodbye | +3.580 |
Neg logit contributions | |
ibaba | -3.860 |
racuse | -3.704 |
WARE | -3.620 |
buquerque | -3.610 |
Vector | -3.594 |
Token | slump |
Token ID | 37758 |
Feature activation | +4.89e+00 |
Pos logit contributions | |
arthed | +2.888 |
ept | +2.758 |
##### | +2.696 |
Dol | +2.652 |
iltr | +2.617 |
Neg logit contributions | |
ibaba | -2.592 |
racuse | -2.506 |
Vector | -2.446 |
WARE | -2.439 |
buquerque | -2.415 |
Token | in |
Token ID | 287 |
Feature activation | +6.54e+00 |
Pos logit contributions | |
arthed | +3.364 |
ept | +3.170 |
##### | +3.092 |
Dol | +3.063 |
goodbye | +3.044 |
Neg logit contributions | |
ibaba | -3.128 |
racuse | -3.062 |
Vector | -2.950 |
buquerque | -2.926 |
WARE | -2.903 |
Token | sales |
Token ID | 4200 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
arthed | +4.290 |
ept | +4.112 |
Dol | +3.989 |
##### | +3.956 |
goodbye | +3.904 |
Neg logit contributions | |
ibaba | -4.247 |
racuse | -4.150 |
WARE | -4.054 |
buquerque | -4.012 |
Vector | -4.010 |
Token | over |
Token ID | 625 |
Feature activation | +1.14e+01 |
Pos logit contributions | |
arthed | +2.346 |
ept | +2.223 |
##### | +2.167 |
Dol | +2.120 |
goodbye | +2.106 |
Neg logit contributions | |
ibaba | -2.080 |
racuse | -2.011 |
Vector | -1.978 |
WARE | -1.935 |
buquerque | -1.932 |
Token | the |
Token ID | 262 |
Feature activation | +5.68e+00 |
Pos logit contributions | |
arthed | +6.351 |
ept | +5.937 |
##### | +5.761 |
Dol | +5.678 |
goodbye | +5.651 |
Neg logit contributions | |
ibaba | -7.883 |
racuse | -7.879 |
WARE | -7.679 |
Vector | -7.477 |
ABLE | -7.475 |
Token | cold |
Token ID | 4692 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
arthed | +3.276 |
##### | +3.020 |
ept | +3.004 |
Dol | +2.948 |
goodbye | +2.932 |
Neg logit contributions | |
ibaba | -4.211 |
racuse | -4.181 |
WARE | -4.004 |
Vector | -3.979 |
buquerque | -3.940 |
Token | winter |
Token ID | 7374 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
arthed | +0.099 |
ept | +0.092 |
##### | +0.090 |
iltr | +0.088 |
Dol | +0.086 |
Neg logit contributions | |
ibaba | -0.097 |
Vector | -0.094 |
racuse | -0.093 |
WARE | -0.091 |
buquerque | -0.090 |
Token | months |
Token ID | 1933 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
ibaba | +1.036 |
Vector | +1.002 |
racuse | +0.971 |
WARE | +0.968 |
buquerque | +0.954 |
Neg logit contributions | |
arthed | -1.152 |
ept | -1.071 |
##### | -1.045 |
iltr | -1.031 |
Dol | -0.994 |
Token | hasn |
Token ID | 5818 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
ibaba | +0.179 |
Vector | +0.173 |
racuse | +0.170 |
WARE | +0.166 |
buquerque | +0.165 |
Neg logit contributions | |
arthed | -0.224 |
ept | -0.210 |
##### | -0.205 |
iltr | -0.202 |
Dol | -0.197 |
Token | in |
Token ID | 287 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
arthed | +2.453 |
ept | +2.312 |
##### | +2.256 |
iltr | +2.200 |
goodbye | +2.193 |
Neg logit contributions | |
ibaba | -2.295 |
Vector | -2.188 |
racuse | -2.183 |
WARE | -2.126 |
buquerque | -2.125 |
Token | Game |
Token ID | 3776 |
Feature activation | +4.47e+00 |
Pos logit contributions | |
ibaba | +1.438 |
Vector | +1.402 |
racuse | +1.342 |
Higgins | +1.314 |
buquerque | +1.312 |
Neg logit contributions | |
arthed | -2.273 |
ept | -2.113 |
##### | -2.066 |
iltr | -2.064 |
Dol | -1.977 |
Token | 4 |
Token ID | 604 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
arthed | +3.008 |
##### | +2.853 |
ept | +2.824 |
Dol | +2.732 |
iltr | +2.686 |
Neg logit contributions | |
ibaba | -2.941 |
racuse | -2.841 |
WARE | -2.819 |
Vector | -2.762 |
buquerque | -2.728 |
Token | of |
Token ID | 286 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
arthed | +1.426 |
ept | +1.341 |
##### | +1.307 |
iltr | +1.269 |
goodbye | +1.265 |
Neg logit contributions | |
ibaba | -1.344 |
Vector | -1.288 |
racuse | -1.280 |
WARE | -1.252 |
buquerque | -1.240 |
Token | an |
Token ID | 281 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
arthed | +2.021 |
ept | +1.892 |
##### | +1.850 |
goodbye | +1.797 |
Dol | +1.796 |
Neg logit contributions | |
ibaba | -1.921 |
racuse | -1.844 |
Vector | -1.841 |
WARE | -1.833 |
buquerque | -1.772 |
Token | Eastern |
Token ID | 8345 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
arthed | +0.780 |
ept | +0.719 |
##### | +0.696 |
iltr | +0.677 |
Dol | +0.661 |
Neg logit contributions | |
ibaba | -1.172 |
racuse | -1.134 |
Vector | -1.131 |
WARE | -1.114 |
buquerque | -1.101 |
Token | Conference |
Token ID | 8785 |
Feature activation | +6.31e+00 |
Pos logit contributions | |
arthed | +5.196 |
ept | +4.883 |
##### | +4.688 |
Dol | +4.655 |
goodbye | +4.538 |
Neg logit contributions | |
ibaba | -8.312 |
buquerque | -8.043 |
WARE | -8.036 |
racuse | -7.981 |
Vector | -7.890 |
Token | playoff |
Token ID | 13119 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
arthed | +3.730 |
ept | +3.470 |
##### | +3.399 |
goodbye | +3.361 |
Dol | +3.339 |
Neg logit contributions | |
ibaba | -4.570 |
racuse | -4.537 |
WARE | -4.405 |
Vector | -4.404 |
buquerque | -4.400 |
Token | series |
Token ID | 2168 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
ibaba | +0.211 |
Vector | +0.203 |
racuse | +0.199 |
WARE | +0.195 |
buquerque | +0.193 |
Neg logit contributions | |
arthed | -0.287 |
ept | -0.269 |
##### | -0.263 |
iltr | -0.259 |
Dol | -0.253 |
Token | , |
Token ID | 11 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
ibaba | +0.656 |
Vector | +0.632 |
racuse | +0.621 |
WARE | +0.608 |
buquerque | +0.606 |
Neg logit contributions | |
arthed | -0.810 |
ept | -0.757 |
##### | -0.740 |
iltr | -0.727 |
Dol | -0.707 |
Token | said |
Token ID | 531 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
ibaba | +1.562 |
Vector | +1.525 |
racuse | +1.471 |
Higgins | +1.442 |
WARE | +1.429 |
Neg logit contributions | |
arthed | -2.161 |
ept | -2.023 |
##### | -1.964 |
iltr | -1.948 |
Dol | -1.873 |
Token | is |
Token ID | 318 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
ibaba | +1.103 |
Vector | +1.068 |
racuse | +1.037 |
WARE | +1.028 |
buquerque | +1.015 |
Neg logit contributions | |
arthed | -1.389 |
ept | -1.289 |
##### | -1.261 |
iltr | -1.243 |
goodbye | -1.200 |
Token | subtle |
Token ID | 11800 |
Feature activation | +3.17e+00 |
Pos logit contributions | |
arthed | +6.519 |
Dol | +6.390 |
ept | +6.346 |
##### | +6.284 |
goodbye | +6.121 |
Neg logit contributions | |
ibaba | -7.616 |
racuse | -7.380 |
buquerque | -7.027 |
Vector | -6.935 |
WARE | -6.918 |
Token | , |
Token ID | 11 |
Feature activation | +4.08e+00 |
Pos logit contributions | |
arthed | +2.239 |
ept | +2.125 |
##### | +2.082 |
iltr | +2.016 |
Dol | +2.014 |
Neg logit contributions | |
ibaba | -2.025 |
Vector | -1.930 |
racuse | -1.927 |
WARE | -1.868 |
buquerque | -1.849 |
Token | but |
Token ID | 475 |
Feature activation | +4.83e+00 |
Pos logit contributions | |
arthed | +2.757 |
ept | +2.608 |
##### | +2.551 |
Dol | +2.498 |
goodbye | +2.466 |
Neg logit contributions | |
ibaba | -2.726 |
racuse | -2.595 |
Vector | -2.558 |
WARE | -2.505 |
buquerque | -2.495 |
Token | every |
Token ID | 790 |
Feature activation | +1.43e+01 |
Pos logit contributions | |
arthed | +3.214 |
ept | +3.056 |
##### | +2.976 |
Dol | +2.937 |
goodbye | +2.889 |
Neg logit contributions | |
ibaba | -3.257 |
racuse | -3.132 |
Vector | -3.045 |
buquerque | -2.990 |
WARE | -2.967 |
Token | few |
Token ID | 1178 |
Feature activation | +1.32e+01 |
Pos logit contributions | |
ept | +7.511 |
Dol | +7.467 |
goodbye | +7.441 |
arthed | +7.388 |
##### | +7.167 |
Neg logit contributions | |
ibaba | -10.078 |
WARE | -9.408 |
racuse | -9.393 |
buquerque | -9.140 |
Higgins | -8.830 |
Token | bites |
Token ID | 26081 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
ept | +6.459 |
arthed | +6.328 |
Dol | +6.214 |
##### | +6.021 |
goodbye | +6.012 |
Neg logit contributions | |
ibaba | -10.125 |
racuse | -9.878 |
buquerque | -9.468 |
Vector | -9.322 |
WARE | -9.250 |
Token | it |
Token ID | 340 |
Feature activation | +7.69e+00 |
Pos logit contributions | |
arthed | +2.853 |
ept | +2.693 |
##### | +2.651 |
Dol | +2.591 |
goodbye | +2.584 |
Neg logit contributions | |
ibaba | -2.650 |
racuse | -2.525 |
Vector | -2.495 |
WARE | -2.438 |
buquerque | -2.431 |
Token | makes |
Token ID | 1838 |
Feature activation | +5.38e+00 |
Pos logit contributions | |
arthed | +4.926 |
ept | +4.855 |
##### | +4.795 |
goodbye | +4.681 |
Dol | +4.666 |
Neg logit contributions | |
ibaba | -5.065 |
racuse | -4.843 |
WARE | -4.804 |
Vector | -4.769 |
buquerque | -4.683 |
Token | its |
Token ID | 663 |
Feature activation | +4.76e+00 |
Pos logit contributions | |
arthed | +3.493 |
ept | +3.298 |
##### | +3.266 |
goodbye | +3.205 |
Dol | +3.201 |
Neg logit contributions | |
ibaba | -3.647 |
racuse | -3.450 |
Vector | -3.411 |
WARE | -3.408 |
buquerque | -3.332 |
Token | presence |
Token ID | 4931 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
arthed | +3.088 |
ept | +2.922 |
##### | +2.857 |
Dol | +2.814 |
goodbye | +2.795 |
Neg logit contributions | |
ibaba | -3.290 |
racuse | -3.138 |
Vector | -3.073 |
WARE | -3.005 |
buquerque | -2.984 |
Token | , |
Token ID | 11 |
Feature activation | +6.81e+00 |
Pos logit contributions | |
arthed | +2.535 |
ept | +2.377 |
##### | +2.353 |
iltr | +2.268 |
Dol | +2.227 |
Neg logit contributions | |
ibaba | -2.337 |
Vector | -2.193 |
racuse | -2.166 |
WARE | -2.140 |
buquerque | -2.114 |
Token | 1 |
Token ID | 352 |
Feature activation | +7.63e-01 |
Pos logit contributions | |
arthed | +4.374 |
ept | +4.172 |
##### | +4.134 |
iltr | +3.927 |
goodbye | +3.889 |
Neg logit contributions | |
ibaba | -4.541 |
racuse | -4.319 |
WARE | -4.263 |
buquerque | -4.194 |
Vector | -4.158 |
Token | , |
Token ID | 11 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
arthed | +0.564 |
ept | +0.530 |
##### | +0.518 |
iltr | +0.507 |
Dol | +0.497 |
Neg logit contributions | |
ibaba | -0.468 |
Vector | -0.446 |
racuse | -0.441 |
WARE | -0.433 |
buquerque | -0.428 |
Token | Length |
Token ID | 22313 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
arthed | +0.275 |
ept | +0.258 |
##### | +0.252 |
iltr | +0.247 |
Dol | +0.241 |
Neg logit contributions | |
ibaba | -0.242 |
Vector | -0.231 |
racuse | -0.229 |
WARE | -0.225 |
buquerque | -0.223 |
Token | [ |
Token ID | 58 |
Feature activation | +6.33e+00 |
Pos logit contributions | |
arthed | +0.593 |
ept | +0.546 |
##### | +0.532 |
iltr | +0.516 |
Dol | +0.501 |
Neg logit contributions | |
ibaba | -0.782 |
Vector | -0.752 |
racuse | -0.745 |
WARE | -0.733 |
buquerque | -0.729 |
Token | Keys |
Token ID | 40729 |
Feature activation | +1.23e+01 |
Pos logit contributions | |
arthed | +4.109 |
ept | +3.932 |
##### | +3.863 |
iltr | +3.733 |
goodbye | +3.615 |
Neg logit contributions | |
ibaba | -4.212 |
racuse | -4.018 |
Vector | -3.932 |
WARE | -3.871 |
buquerque | -3.844 |
Token | [ |
Token ID | 58 |
Feature activation | +7.07e+00 |
Pos logit contributions | |
arthed | +4.376 |
##### | +4.374 |
ept | +4.065 |
/# | +3.954 |
ⓘ | +3.819 |
Neg logit contributions | |
ibaba | -11.786 |
ccording | -11.596 |
buquerque | -10.904 |
racuse | -10.821 |
Vector | -10.488 |
Token | search |
Token ID | 12947 |
Feature activation | +4.33e+00 |
Pos logit contributions | |
arthed | +5.300 |
ept | +5.012 |
##### | +5.006 |
iltr | +4.863 |
goodbye | +4.659 |
Neg logit contributions | |
ibaba | -3.929 |
racuse | -3.827 |
Vector | -3.768 |
WARE | -3.677 |
Higgins | -3.656 |
Token | Space |
Token ID | 14106 |
Feature activation | +1.13e+01 |
Pos logit contributions | |
arthed | +2.496 |
ept | +2.332 |
##### | +2.294 |
iltr | +2.170 |
ⓘ | +2.142 |
Neg logit contributions | |
ibaba | -3.418 |
racuse | -3.208 |
buquerque | -3.138 |
Vector | -3.110 |
Higgins | -3.092 |
Token | ]] |
Token ID | 11907 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
arthed | +4.672 |
ept | +4.232 |
##### | +4.190 |
/# | +3.819 |
iltr | +3.778 |
Neg logit contributions | |
ibaba | -9.815 |
WARE | -9.394 |
Vector | -9.368 |
racuse | -9.245 |
Notting | -9.206 |
Token | } |
Token ID | 92 |
Feature activation | +3.97e+00 |
Pos logit contributions | |
arthed | +1.071 |
ept | +1.005 |
##### | +0.985 |
iltr | +0.957 |
Dol | +0.940 |
Neg logit contributions | |
ibaba | -0.935 |
Vector | -0.889 |
racuse | -0.883 |
WARE | -0.864 |
buquerque | -0.858 |
Token | , |
Token ID | 11 |
Feature activation | -7.80e-02 |
Pos logit contributions | |
ibaba | +2.047 |
Vector | +2.003 |
racuse | +1.904 |
Higgins | +1.862 |
WARE | +1.861 |
Neg logit contributions | |
arthed | -2.878 |
ept | -2.646 |
##### | -2.604 |
iltr | -2.562 |
goodbye | -2.497 |
Token | Eve |
Token ID | 12882 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ibaba | +0.049 |
Vector | +0.047 |
racuse | +0.046 |
WARE | +0.045 |
buquerque | +0.045 |
Neg logit contributions | |
arthed | -0.057 |
ept | -0.053 |
##### | -0.052 |
iltr | -0.051 |
Dol | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
arthed | +0.854 |
ept | +0.804 |
##### | +0.781 |
iltr | +0.766 |
Dol | +0.754 |
Neg logit contributions | |
ibaba | -0.702 |
Vector | -0.675 |
racuse | -0.669 |
WARE | -0.655 |
buquerque | -0.647 |
Token | other |
Token ID | 584 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
arthed | +2.033 |
ept | +1.931 |
##### | +1.890 |
Dol | +1.856 |
iltr | +1.834 |
Neg logit contributions | |
ibaba | -1.635 |
racuse | -1.547 |
WARE | -1.528 |
Vector | -1.521 |
buquerque | -1.509 |
Token | cigarette |
Token ID | 17779 |
Feature activation | -7.30e+00 |
Pos logit contributions | |
arthed | +1.343 |
ept | +1.273 |
##### | +1.238 |
Dol | +1.210 |
iltr | +1.208 |
Neg logit contributions | |
ibaba | -1.276 |
racuse | -1.213 |
Vector | -1.207 |
WARE | -1.200 |
buquerque | -1.196 |
Token | brands |
Token ID | 11394 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
ibaba | +4.081 |
Vector | +4.012 |
Higgins | +3.878 |
ABLE | +3.829 |
racuse | +3.827 |
Neg logit contributions | |
arthed | -5.526 |
ept | -5.096 |
##### | -5.004 |
iltr | -4.982 |
ⓘ | -4.656 |
Token | , |
Token ID | 11 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
arthed | +0.098 |
ept | +0.091 |
##### | +0.089 |
iltr | +0.087 |
Dol | +0.086 |
Neg logit contributions | |
ibaba | -0.085 |
Vector | -0.081 |
racuse | -0.080 |
WARE | -0.079 |
buquerque | -0.078 |
Token | saw |
Token ID | 2497 |
Feature activation | -7.00e+00 |
Pos logit contributions | |
ibaba | +0.750 |
Vector | +0.728 |
racuse | +0.711 |
WARE | +0.693 |
buquerque | +0.688 |
Neg logit contributions | |
arthed | -0.918 |
ept | -0.856 |
##### | -0.835 |
iltr | -0.823 |
goodbye | -0.797 |
Token | its |
Token ID | 663 |
Feature activation | -5.72e+00 |
Pos logit contributions | |
ibaba | +3.994 |
Vector | +3.933 |
Higgins | +3.772 |
racuse | +3.746 |
Startup | +3.650 |
Neg logit contributions | |
arthed | -5.394 |
ept | -4.979 |
iltr | -4.881 |
##### | -4.825 |
ⓘ | -4.503 |
Token | stock |
Token ID | 4283 |
Feature activation | -3.78e+00 |
Pos logit contributions | |
ibaba | +3.069 |
Vector | +3.037 |
racuse | +2.877 |
Higgins | +2.869 |
Startup | +2.836 |
Neg logit contributions | |
arthed | -4.686 |
ept | -4.281 |
iltr | -4.174 |
##### | -4.132 |
goodbye | -3.934 |
Token | decline |
Token ID | 7794 |
Feature activation | +6.41e-02 |
Pos logit contributions | |
ibaba | +2.102 |
Vector | +2.037 |
racuse | +1.960 |
WARE | +1.912 |
Higgins | +1.903 |
Neg logit contributions | |
arthed | -3.064 |
ept | -2.804 |
##### | -2.731 |
iltr | -2.720 |
Dol | -2.613 |
Token | Liberia |
Token ID | 33208 |
Feature activation | +6.66e+00 |
Pos logit contributions | |
ibaba | +1.285 |
Vector | +1.251 |
racuse | +1.201 |
buquerque | +1.195 |
Higgins | +1.179 |
Neg logit contributions | |
arthed | -1.803 |
ept | -1.676 |
##### | -1.644 |
iltr | -1.626 |
goodbye | -1.575 |
Token | , |
Token ID | 11 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
arthed | +4.336 |
ept | +4.064 |
##### | +4.010 |
goodbye | +3.902 |
Dol | +3.849 |
Neg logit contributions | |
ibaba | -4.469 |
racuse | -4.332 |
WARE | -4.205 |
Vector | -4.173 |
buquerque | -4.073 |
Token | but |
Token ID | 475 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
arthed | +0.148 |
ept | +0.138 |
##### | +0.135 |
iltr | +0.133 |
Dol | +0.130 |
Neg logit contributions | |
ibaba | -0.118 |
Vector | -0.113 |
racuse | -0.112 |
WARE | -0.109 |
buquerque | -0.108 |
Token | was |
Token ID | 373 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
ibaba | +0.296 |
Vector | +0.285 |
racuse | +0.280 |
WARE | +0.275 |
buquerque | +0.273 |
Neg logit contributions | |
arthed | -0.356 |
ept | -0.333 |
##### | -0.325 |
iltr | -0.320 |
Dol | -0.312 |
Token | as |
Token ID | 355 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
arthed | +0.765 |
ept | +0.717 |
##### | +0.703 |
iltr | +0.686 |
Dol | +0.675 |
Neg logit contributions | |
ibaba | -0.665 |
Vector | -0.631 |
racuse | -0.630 |
WARE | -0.616 |
buquerque | -0.609 |
Token | ym |
Token ID | 4948 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
arthed | +1.965 |
ept | +1.863 |
##### | +1.817 |
iltr | +1.771 |
Dol | +1.755 |
Neg logit contributions | |
ibaba | -1.323 |
racuse | -1.255 |
Vector | -1.232 |
WARE | -1.196 |
buquerque | -1.180 |
Token | pt |
Token ID | 457 |
Feature activation | +6.36e+00 |
Pos logit contributions | |
ibaba | +0.157 |
Vector | +0.150 |
racuse | +0.147 |
WARE | +0.143 |
buquerque | +0.142 |
Neg logit contributions | |
arthed | -0.284 |
ept | -0.269 |
##### | -0.264 |
iltr | -0.260 |
Dol | -0.255 |
Token | omatic |
Token ID | 13730 |
Feature activation | +8.24e-01 |
Pos logit contributions | |
arthed | +3.503 |
ept | +3.262 |
iltr | +3.191 |
##### | +3.075 |
bell | +2.893 |
Neg logit contributions | |
ibaba | -5.070 |
Vector | -4.974 |
WARE | -4.963 |
racuse | -4.769 |
Higgins | -4.713 |
Token | until |
Token ID | 1566 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
arthed | +0.589 |
ept | +0.552 |
##### | +0.539 |
iltr | +0.529 |
Dol | +0.518 |
Neg logit contributions | |
ibaba | -0.524 |
Vector | -0.503 |
racuse | -0.499 |
WARE | -0.488 |
buquerque | -0.483 |
Token | after |
Token ID | 706 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
ibaba | +0.141 |
Vector | +0.136 |
racuse | +0.134 |
WARE | +0.131 |
buquerque | +0.131 |
Neg logit contributions | |
arthed | -0.148 |
ept | -0.138 |
##### | -0.135 |
iltr | -0.132 |
Dol | -0.129 |
Token | his |
Token ID | 465 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
arthed | +1.119 |
ept | +1.043 |
##### | +1.019 |
iltr | +0.995 |
goodbye | +0.982 |
Neg logit contributions | |
ibaba | -1.141 |
racuse | -1.087 |
Vector | -1.083 |
WARE | -1.063 |
buquerque | -1.050 |
Token | for |
Token ID | 329 |
Feature activation | +4.31e+00 |
Pos logit contributions | |
ibaba | +0.414 |
Vector | +0.400 |
racuse | +0.390 |
WARE | +0.384 |
buquerque | +0.381 |
Neg logit contributions | |
arthed | -0.514 |
ept | -0.481 |
##### | -0.469 |
iltr | -0.462 |
Dol | -0.449 |
Token | the |
Token ID | 262 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
arthed | +2.903 |
ept | +2.744 |
##### | +2.738 |
Dol | +2.671 |
goodbye | +2.627 |
Neg logit contributions | |
ibaba | -2.839 |
racuse | -2.736 |
WARE | -2.656 |
Vector | -2.652 |
buquerque | -2.623 |
Token | Syrian |
Token ID | 6318 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
ibaba | +0.995 |
Vector | +0.970 |
racuse | +0.939 |
WARE | +0.925 |
buquerque | +0.917 |
Neg logit contributions | |
arthed | -1.273 |
ept | -1.185 |
##### | -1.149 |
iltr | -1.141 |
Dol | -1.100 |
Token | regime |
Token ID | 7142 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
arthed | +1.689 |
ept | +1.600 |
##### | +1.578 |
iltr | +1.531 |
Dol | +1.521 |
Neg logit contributions | |
ibaba | -1.413 |
racuse | -1.345 |
Vector | -1.330 |
WARE | -1.301 |
buquerque | -1.291 |
Token | , |
Token ID | 11 |
Feature activation | -5.36e-02 |
Pos logit contributions | |
ibaba | +1.338 |
Vector | +1.297 |
racuse | +1.249 |
WARE | +1.238 |
buquerque | +1.228 |
Neg logit contributions | |
arthed | -1.762 |
ept | -1.641 |
##### | -1.585 |
iltr | -1.573 |
Dol | -1.523 |
Token | to |
Token ID | 284 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
ibaba | +0.034 |
Vector | +0.033 |
racuse | +0.033 |
WARE | +0.032 |
buquerque | +0.032 |
Neg logit contributions | |
arthed | -0.038 |
ept | -0.036 |
##### | -0.035 |
iltr | -0.034 |
Dol | -0.033 |
Token | have |
Token ID | 423 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
arthed | +1.991 |
ept | +1.868 |
##### | +1.854 |
iltr | +1.799 |
Dol | +1.788 |
Neg logit contributions | |
ibaba | -1.975 |
racuse | -1.908 |
Vector | -1.874 |
WARE | -1.844 |
buquerque | -1.838 |
Token | a |
Token ID | 257 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
ibaba | +1.695 |
Vector | +1.672 |
racuse | +1.593 |
Higgins | +1.586 |
WARE | +1.575 |
Neg logit contributions | |
arthed | -2.051 |
ept | -1.912 |
##### | -1.847 |
iltr | -1.837 |
Dol | -1.753 |
Token | Syrian |
Token ID | 6318 |
Feature activation | +3.83e+00 |
Pos logit contributions | |
arthed | +1.540 |
ept | +1.449 |
##### | +1.435 |
iltr | +1.384 |
Dol | +1.382 |
Neg logit contributions | |
ibaba | -1.383 |
racuse | -1.317 |
Vector | -1.302 |
WARE | -1.283 |
buquerque | -1.273 |
Token | Benghazi |
Token ID | 22497 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
arthed | +2.668 |
ept | +2.519 |
##### | +2.496 |
Dol | +2.424 |
iltr | +2.404 |
Neg logit contributions | |
ibaba | -2.486 |
racuse | -2.368 |
WARE | -2.305 |
Vector | -2.301 |
buquerque | -2.274 |
Token | , |
Token ID | 11 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
arthed | +1.451 |
ept | +1.354 |
##### | +1.333 |
iltr | +1.299 |
Dol | +1.286 |
Neg logit contributions | |
ibaba | -1.313 |
racuse | -1.251 |
Vector | -1.246 |
WARE | -1.226 |
buquerque | -1.209 |
Token | . |
Token ID | 13 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
arthed | +1.097 |
ept | +1.033 |
##### | +1.011 |
iltr | +0.989 |
goodbye | +0.972 |
Neg logit contributions | |
ibaba | -0.967 |
Vector | -0.925 |
racuse | -0.914 |
WARE | -0.899 |
buquerque | -0.885 |
Token | That |
Token ID | 1320 |
Feature activation | -6.16e+00 |
Pos logit contributions | |
ibaba | +0.304 |
Vector | +0.294 |
racuse | +0.288 |
WARE | +0.282 |
buquerque | +0.280 |
Neg logit contributions | |
arthed | -0.401 |
ept | -0.376 |
##### | -0.367 |
iltr | -0.361 |
Dol | -0.352 |
Token | didn |
Token ID | 1422 |
Feature activation | -8.29e-01 |
Pos logit contributions | |
ibaba | +3.449 |
Vector | +3.356 |
Higgins | +3.265 |
buquerque | +3.194 |
racuse | +3.159 |
Neg logit contributions | |
arthed | -4.840 |
ept | -4.399 |
##### | -4.264 |
iltr | -4.261 |
goodbye | -4.052 |
Token | � |
Token ID | 447 |
Feature activation | -5.78e+00 |
Pos logit contributions | |
ibaba | +0.571 |
Vector | +0.556 |
racuse | +0.545 |
WARE | +0.537 |
buquerque | +0.532 |
Neg logit contributions | |
arthed | -0.548 |
ept | -0.507 |
##### | -0.491 |
iltr | -0.486 |
Dol | -0.470 |
Token | � |
Token ID | 247 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
ibaba | +2.102 |
Vector | +1.927 |
ABLE | +1.845 |
WARE | +1.844 |
racuse | +1.785 |
Neg logit contributions | |
arthed | -5.541 |
ept | -5.216 |
iltr | -5.214 |
##### | -5.107 |
Dol | -5.072 |
Token | t |
Token ID | 83 |
Feature activation | +7.73e+00 |
Pos logit contributions | |
ibaba | +1.801 |
Vector | +1.745 |
racuse | +1.714 |
WARE | +1.699 |
buquerque | +1.688 |
Neg logit contributions | |
arthed | -1.256 |
ept | -1.122 |
##### | -1.100 |
iltr | -1.080 |
goodbye | -1.041 |
Token | happen |
Token ID | 1645 |
Feature activation | -4.10e+00 |
Pos logit contributions | |
arthed | +4.098 |
ept | +4.054 |
##### | +4.011 |
iltr | +3.846 |
Dol | +3.808 |
Neg logit contributions | |
ibaba | -5.910 |
racuse | -5.819 |
Vector | -5.519 |
WARE | -5.506 |
buquerque | -5.442 |
Token | . |
Token ID | 13 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
ibaba | +2.313 |
Vector | +2.261 |
racuse | +2.172 |
WARE | +2.152 |
buquerque | +2.148 |
Neg logit contributions | |
arthed | -3.200 |
ept | -2.946 |
##### | -2.870 |
iltr | -2.830 |
Dol | -2.747 |
Token | � |
Token ID | 447 |
Feature activation | -4.05e-01 |
Pos logit contributions | |
arthed | +1.637 |
ept | +1.543 |
##### | +1.535 |
iltr | +1.475 |
Dol | +1.473 |
Neg logit contributions | |
ibaba | -1.416 |
racuse | -1.339 |
Vector | -1.324 |
WARE | -1.299 |
buquerque | -1.289 |
Token | � |
Token ID | 251 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
ibaba | +0.170 |
Vector | +0.161 |
racuse | +0.157 |
WARE | +0.153 |
buquerque | +0.151 |
Neg logit contributions | |
arthed | -0.376 |
ept | -0.357 |
##### | -0.351 |
iltr | -0.346 |
Dol | -0.340 |
Token | He |
Token ID | 679 |
Feature activation | +4.51e+00 |
Pos logit contributions | |
arthed | +1.852 |
##### | +1.741 |
ept | +1.741 |
Dol | +1.661 |
iltr | +1.659 |
Neg logit contributions | |
ibaba | -1.795 |
racuse | -1.692 |
Vector | -1.676 |
buquerque | -1.643 |
WARE | -1.642 |
Token | adopted |
Token ID | 8197 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
arthed | +1.431 |
ept | +1.343 |
##### | +1.315 |
iltr | +1.289 |
Dol | +1.278 |
Neg logit contributions | |
ibaba | -1.129 |
Vector | -1.067 |
racuse | -1.061 |
WARE | -1.036 |
buquerque | -1.031 |
Token | by |
Token ID | 416 |
Feature activation | +4.08e+00 |
Pos logit contributions | |
ibaba | +1.527 |
Vector | +1.494 |
racuse | +1.446 |
WARE | +1.429 |
buquerque | +1.415 |
Neg logit contributions | |
arthed | -1.966 |
ept | -1.820 |
##### | -1.780 |
iltr | -1.747 |
Dol | -1.689 |
Token | feminists |
Token ID | 22742 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
arthed | +2.585 |
ept | +2.482 |
##### | +2.444 |
Dol | +2.395 |
iltr | +2.338 |
Neg logit contributions | |
ibaba | -2.866 |
racuse | -2.727 |
Vector | -2.665 |
WARE | -2.655 |
buquerque | -2.620 |
Token | . |
Token ID | 13 |
Feature activation | -6.78e-01 |
Pos logit contributions | |
arthed | +2.694 |
ept | +2.583 |
##### | +2.523 |
Dol | +2.465 |
goodbye | +2.453 |
Neg logit contributions | |
ibaba | -2.827 |
racuse | -2.668 |
Vector | -2.635 |
WARE | -2.575 |
buquerque | -2.574 |
Token | Her |
Token ID | 2332 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
ibaba | +0.401 |
Vector | +0.388 |
racuse | +0.378 |
WARE | +0.371 |
buquerque | +0.369 |
Neg logit contributions | |
arthed | -0.515 |
ept | -0.482 |
##### | -0.470 |
iltr | -0.463 |
goodbye | -0.451 |
Token | argument |
Token ID | 4578 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
arthed | +0.730 |
ept | +0.687 |
##### | +0.671 |
iltr | +0.657 |
Dol | +0.646 |
Neg logit contributions | |
ibaba | -0.639 |
Vector | -0.610 |
racuse | -0.606 |
WARE | -0.593 |
buquerque | -0.588 |
Token | is |
Token ID | 318 |
Feature activation | -4.85e+00 |
Pos logit contributions | |
arthed | +2.551 |
ept | +2.439 |
##### | +2.381 |
iltr | +2.323 |
goodbye | +2.290 |
Neg logit contributions | |
ibaba | -2.224 |
racuse | -2.143 |
Vector | -2.131 |
WARE | -2.099 |
buquerque | -2.053 |
Token | not |
Token ID | 407 |
Feature activation | -9.25e+00 |
Pos logit contributions | |
ibaba | +2.756 |
Vector | +2.745 |
Higgins | +2.605 |
WARE | +2.599 |
ABLE | +2.576 |
Neg logit contributions | |
arthed | -3.705 |
ept | -3.425 |
iltr | -3.325 |
##### | -3.322 |
goodbye | -3.150 |
Token | convincing |
Token ID | 17101 |
Feature activation | -3.55e+00 |
Pos logit contributions | |
Vector | +4.977 |
ibaba | +4.837 |
ABLE | +4.780 |
Higgins | +4.735 |
Hamilton | +4.697 |
Neg logit contributions | |
arthed | -7.038 |
ept | -6.384 |
iltr | -6.294 |
##### | -6.182 |
irs | -5.875 |
Token | as |
Token ID | 355 |
Feature activation | -7.59e+00 |
Pos logit contributions | |
ibaba | +2.068 |
Vector | +2.026 |
WARE | +1.943 |
ABLE | +1.924 |
racuse | +1.918 |
Neg logit contributions | |
arthed | -2.714 |
ept | -2.490 |
##### | -2.439 |
iltr | -2.407 |
Dol | -2.314 |
Token | it |
Token ID | 340 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
ibaba | +4.230 |
Vector | +4.176 |
Higgins | +4.033 |
Startup | +3.894 |
buquerque | +3.878 |
Neg logit contributions | |
arthed | -5.801 |
ept | -5.288 |
iltr | -5.135 |
##### | -5.120 |
goodbye | -4.941 |
Token | o |
Token ID | 267 |
Feature activation | -4.86e+00 |
Pos logit contributions | |
ibaba | +0.917 |
Vector | +0.882 |
racuse | +0.858 |
WARE | +0.843 |
buquerque | +0.840 |
Neg logit contributions | |
arthed | -1.209 |
ept | -1.128 |
##### | -1.105 |
iltr | -1.089 |
Dol | -1.057 |
Token | Hal |
Token ID | 11023 |
Feature activation | -5.41e+00 |
Pos logit contributions | |
ibaba | +2.887 |
Vector | +2.803 |
racuse | +2.665 |
buquerque | +2.650 |
Higgins | +2.622 |
Neg logit contributions | |
arthed | -3.625 |
ept | -3.363 |
##### | -3.298 |
iltr | -3.271 |
goodbye | -3.135 |
Token | ep |
Token ID | 538 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
ibaba | +2.487 |
Vector | +2.250 |
buquerque | +2.234 |
racuse | +2.182 |
Startup | +2.082 |
Neg logit contributions | |
arthed | -4.807 |
##### | -4.467 |
ept | -4.456 |
iltr | -4.375 |
goodbye | -4.351 |
Token | ā |
Token ID | 10235 |
Feature activation | +9.54e-01 |
Pos logit contributions | |
ibaba | +1.157 |
Vector | +1.075 |
racuse | +1.055 |
buquerque | +1.032 |
WARE | +0.999 |
Neg logit contributions | |
arthed | -2.229 |
ept | -2.102 |
##### | -2.075 |
iltr | -2.035 |
Dol | -2.002 |
Token | k |
Token ID | 74 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
arthed | +0.726 |
ept | +0.684 |
##### | +0.667 |
iltr | +0.656 |
Dol | +0.641 |
Neg logit contributions | |
ibaba | -0.562 |
Vector | -0.541 |
racuse | -0.531 |
WARE | -0.522 |
buquerque | -0.515 |
Token | ī |
Token ID | 18962 |
Feature activation | -4.78e+00 |
Pos logit contributions | |
arthed | +2.199 |
ept | +2.100 |
##### | +2.046 |
iltr | +2.032 |
Dol | +1.976 |
Neg logit contributions | |
ibaba | -1.219 |
Vector | -1.176 |
racuse | -1.159 |
WARE | -1.149 |
buquerque | -1.104 |
Token | , |
Token ID | 11 |
Feature activation | -6.87e+00 |
Pos logit contributions | |
ibaba | +2.742 |
Vector | +2.673 |
racuse | +2.557 |
WARE | +2.529 |
buquerque | +2.523 |
Neg logit contributions | |
arthed | -3.686 |
ept | -3.362 |
##### | -3.317 |
iltr | -3.297 |
goodbye | -3.148 |
Token | a |
Token ID | 257 |
Feature activation | -7.64e+00 |
Pos logit contributions | |
ibaba | +3.979 |
Vector | +3.890 |
racuse | +3.674 |
buquerque | +3.650 |
Higgins | +3.616 |
Neg logit contributions | |
arthed | -5.163 |
ept | -4.735 |
##### | -4.614 |
iltr | -4.601 |
CTV | -4.411 |
Token | k |
Token ID | 479 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
ibaba | +4.471 |
Vector | +4.315 |
racuse | +4.118 |
buquerque | +4.097 |
Higgins | +4.033 |
Neg logit contributions | |
arthed | -5.647 |
ept | -5.131 |
##### | -5.095 |
iltr | -5.017 |
CTV | -4.854 |
Token | uko |
Token ID | 29794 |
Feature activation | -5.23e+00 |
Pos logit contributions | |
ibaba | +1.195 |
Vector | +1.141 |
racuse | +1.128 |
buquerque | +1.103 |
WARE | +1.092 |
Neg logit contributions | |
arthed | -1.352 |
ept | -1.249 |
##### | -1.235 |
iltr | -1.200 |
Dol | -1.181 |
Token | i |
Token ID | 1312 |
Feature activation | -8.99e+00 |
Pos logit contributions | |
ibaba | +2.990 |
Vector | +2.880 |
racuse | +2.730 |
buquerque | +2.716 |
WARE | +2.700 |
Neg logit contributions | |
arthed | -4.029 |
ept | -3.718 |
##### | -3.652 |
iltr | -3.613 |
Dol | -3.458 |
Token | wouldn |
Token ID | 3636 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
ibaba | +2.139 |
Vector | +2.112 |
racuse | +2.010 |
WARE | +1.988 |
Higgins | +1.983 |
Neg logit contributions | |
arthed | -2.830 |
ept | -2.601 |
##### | -2.480 |
iltr | -2.476 |
Dol | -2.422 |
Token | � |
Token ID | 447 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
arthed | +2.296 |
ept | +2.163 |
##### | +2.156 |
iltr | +2.056 |
Dol | +2.036 |
Neg logit contributions | |
ibaba | -2.203 |
racuse | -2.091 |
Vector | -2.076 |
buquerque | -2.035 |
WARE | -2.026 |
Token | � |
Token ID | 247 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
ibaba | +0.494 |
Vector | +0.468 |
racuse | +0.455 |
WARE | +0.446 |
buquerque | +0.440 |
Neg logit contributions | |
arthed | -0.956 |
ept | -0.904 |
##### | -0.886 |
iltr | -0.878 |
Dol | -0.862 |
Token | t |
Token ID | 83 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
arthed | +1.225 |
ept | +1.146 |
##### | +1.091 |
iltr | +1.064 |
Dol | +1.022 |
Neg logit contributions | |
ibaba | -2.002 |
Vector | -1.938 |
racuse | -1.930 |
WARE | -1.895 |
buquerque | -1.888 |
Token | mind |
Token ID | 2000 |
Feature activation | -5.98e+00 |
Pos logit contributions | |
arthed | +1.002 |
ept | +0.941 |
##### | +0.928 |
iltr | +0.905 |
Dol | +0.886 |
Neg logit contributions | |
ibaba | -0.904 |
racuse | -0.862 |
Vector | -0.861 |
WARE | -0.837 |
buquerque | -0.832 |
Token | trying |
Token ID | 2111 |
Feature activation | -6.17e+00 |
Pos logit contributions | |
Vector | +3.293 |
ibaba | +3.218 |
Higgins | +3.062 |
Startup | +3.030 |
buquerque | +3.027 |
Neg logit contributions | |
arthed | -4.713 |
ept | -4.378 |
##### | -4.205 |
iltr | -4.173 |
Dol | -4.004 |
Token | this |
Token ID | 428 |
Feature activation | -7.97e+00 |
Pos logit contributions | |
Vector | +3.419 |
ibaba | +3.366 |
Startup | +3.156 |
buquerque | +3.139 |
Higgins | +3.132 |
Neg logit contributions | |
arthed | -4.870 |
ept | -4.521 |
iltr | -4.304 |
##### | -4.251 |
Dol | -4.113 |
Token | out |
Token ID | 503 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
Vector | +4.182 |
ibaba | +4.021 |
Startup | +3.830 |
Higgins | +3.811 |
WARE | +3.734 |
Neg logit contributions | |
arthed | -6.527 |
ept | -5.896 |
iltr | -5.755 |
##### | -5.635 |
brate | -5.453 |
Token | . |
Token ID | 13 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
ibaba | +0.658 |
Vector | +0.637 |
racuse | +0.621 |
WARE | +0.610 |
buquerque | +0.605 |
Neg logit contributions | |
arthed | -0.811 |
ept | -0.756 |
##### | -0.737 |
iltr | -0.726 |
Dol | -0.707 |
Token | I |
Token ID | 314 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
ibaba | +1.224 |
Vector | +1.211 |
racuse | +1.144 |
WARE | +1.139 |
Higgins | +1.135 |
Neg logit contributions | |
arthed | -1.655 |
ept | -1.549 |
##### | -1.490 |
iltr | -1.484 |
goodbye | -1.439 |
Token | can |
Token ID | 460 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
ibaba | +1.312 |
Vector | +1.279 |
racuse | +1.232 |
WARE | +1.217 |
buquerque | +1.207 |
Neg logit contributions | |
arthed | -1.663 |
ept | -1.539 |
##### | -1.485 |
iltr | -1.471 |
Dol | -1.439 |
Token |
|
Token ID | 198 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
arthed | +2.254 |
ept | +2.122 |
##### | +2.113 |
Dol | +2.023 |
iltr | +1.996 |
Neg logit contributions | |
ibaba | -2.366 |
racuse | -2.248 |
buquerque | -2.205 |
Vector | -2.191 |
WARE | -2.173 |
Token | Newsletter |
Token ID | 33031 |
Feature activation | +3.57e+00 |
Pos logit contributions | |
ibaba | +0.974 |
Vector | +0.940 |
racuse | +0.920 |
WARE | +0.915 |
buquerque | +0.899 |
Neg logit contributions | |
arthed | -1.308 |
ept | -1.223 |
##### | -1.200 |
iltr | -1.183 |
Dol | -1.154 |
Token | Sign |
Token ID | 5865 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
arthed | +1.319 |
ept | +1.211 |
##### | +1.195 |
Dol | +1.113 |
iltr | +1.061 |
Neg logit contributions | |
ibaba | -3.469 |
racuse | -3.358 |
buquerque | -3.295 |
WARE | -3.283 |
Vector | -3.274 |
Token | Up |
Token ID | 3205 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
arthed | +1.509 |
ept | +1.434 |
##### | +1.407 |
iltr | +1.364 |
Dol | +1.356 |
Neg logit contributions | |
ibaba | -1.237 |
racuse | -1.162 |
Vector | -1.153 |
WARE | -1.147 |
buquerque | -1.135 |
Token | Continue |
Token ID | 10054 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
ibaba | +1.069 |
Vector | +1.041 |
racuse | +1.013 |
WARE | +0.988 |
buquerque | +0.984 |
Neg logit contributions | |
arthed | -1.605 |
ept | -1.494 |
##### | -1.455 |
iltr | -1.447 |
goodbye | -1.404 |
Token | reading |
Token ID | 3555 |
Feature activation | -1.27e+01 |
Pos logit contributions | |
ibaba | +0.105 |
Vector | +0.099 |
racuse | +0.096 |
WARE | +0.093 |
buquerque | +0.092 |
Neg logit contributions | |
arthed | -0.276 |
ept | -0.263 |
##### | -0.258 |
iltr | -0.255 |
Dol | -0.251 |
Token | the |
Token ID | 262 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
Vector | +4.945 |
Hamilton | +4.555 |
Startup | +4.544 |
Higgins | +4.497 |
ABLE | +4.198 |
Neg logit contributions | |
arthed | -11.468 |
iltr | -11.000 |
ept | -10.808 |
##### | -10.288 |
igne | -10.104 |
Token | main |
Token ID | 1388 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
arthed | +1.748 |
ept | +1.654 |
##### | +1.620 |
goodbye | +1.557 |
Dol | +1.552 |
Neg logit contributions | |
ibaba | -1.908 |
racuse | -1.828 |
Vector | -1.814 |
WARE | -1.801 |
buquerque | -1.797 |
Token | story |
Token ID | 1621 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
arthed | +0.684 |
ept | +0.622 |
##### | +0.604 |
iltr | +0.577 |
Dol | +0.566 |
Neg logit contributions | |
ibaba | -1.242 |
Vector | -1.198 |
racuse | -1.193 |
WARE | -1.184 |
buquerque | -1.172 |
Token | Please |
Token ID | 4222 |
Feature activation | +9.66e-01 |
Pos logit contributions | |
arthed | +1.728 |
ept | +1.633 |
##### | +1.604 |
Dol | +1.553 |
goodbye | +1.529 |
Neg logit contributions | |
ibaba | -1.830 |
racuse | -1.745 |
Vector | -1.723 |
WARE | -1.696 |
buquerque | -1.692 |
Token | verify |
Token ID | 11767 |
Feature activation | +3.36e+00 |
Pos logit contributions | |
arthed | +0.709 |
ept | +0.665 |
##### | +0.653 |
iltr | +0.641 |
goodbye | +0.628 |
Neg logit contributions | |
ibaba | -0.595 |
Vector | -0.567 |
racuse | -0.564 |
WARE | -0.556 |
buquerque | -0.548 |
Token | . |
Token ID | 13 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
arthed | +1.868 |
ept | +1.773 |
##### | +1.749 |
iltr | +1.704 |
Dol | +1.681 |
Neg logit contributions | |
ibaba | -1.730 |
Vector | -1.651 |
racuse | -1.636 |
WARE | -1.612 |
buquerque | -1.577 |
Token | Then |
Token ID | 3244 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
arthed | +1.697 |
ept | +1.607 |
##### | +1.598 |
Dol | +1.544 |
iltr | +1.534 |
Neg logit contributions | |
ibaba | -1.433 |
racuse | -1.363 |
Vector | -1.336 |
WARE | -1.318 |
buquerque | -1.304 |
Token | I |
Token ID | 314 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
arthed | +1.581 |
ept | +1.500 |
##### | +1.472 |
iltr | +1.426 |
Dol | +1.420 |
Neg logit contributions | |
ibaba | -1.457 |
Vector | -1.403 |
racuse | -1.399 |
WARE | -1.369 |
buquerque | -1.351 |
Token | � |
Token ID | 447 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
ibaba | +1.214 |
Vector | +1.178 |
racuse | +1.145 |
WARE | +1.121 |
buquerque | +1.120 |
Neg logit contributions | |
arthed | -1.538 |
ept | -1.423 |
##### | -1.381 |
iltr | -1.363 |
Dol | -1.335 |
Token | � |
Token ID | 247 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
ibaba | +0.021 |
Vector | +0.020 |
racuse | +0.020 |
WARE | +0.019 |
buquerque | +0.019 |
Neg logit contributions | |
arthed | -0.038 |
ept | -0.036 |
##### | -0.036 |
iltr | -0.035 |
Dol | -0.034 |
Token | ll |
Token ID | 297 |
Feature activation | -3.47e+00 |
Pos logit contributions | |
ibaba | +1.660 |
Vector | +1.607 |
racuse | +1.570 |
buquerque | +1.539 |
WARE | +1.537 |
Neg logit contributions | |
arthed | -1.690 |
ept | -1.550 |
##### | -1.521 |
iltr | -1.495 |
goodbye | -1.463 |
Token | say |
Token ID | 910 |
Feature activation | +6.14e+00 |
Pos logit contributions | |
ibaba | +2.030 |
Vector | +1.986 |
racuse | +1.905 |
buquerque | +1.876 |
WARE | +1.870 |
Neg logit contributions | |
arthed | -2.647 |
ept | -2.436 |
##### | -2.356 |
iltr | -2.340 |
Dol | -2.279 |
Token | , |
Token ID | 11 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
arthed | +3.960 |
ept | +3.814 |
goodbye | +3.770 |
##### | +3.765 |
Dol | +3.687 |
Neg logit contributions | |
ibaba | -4.107 |
racuse | -3.949 |
Vector | -3.836 |
WARE | -3.827 |
buquerque | -3.763 |
Token | � |
Token ID | 564 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
arthed | +0.851 |
ept | +0.801 |
##### | +0.784 |
iltr | +0.764 |
Dol | +0.752 |
Neg logit contributions | |
ibaba | -0.811 |
Vector | -0.778 |
racuse | -0.774 |
WARE | -0.758 |
buquerque | -0.751 |
Token | � |
Token ID | 246 |
Feature activation | +8.26e-01 |
Pos logit contributions | |
arthed | +0.924 |
ept | +0.873 |
##### | +0.858 |
iltr | +0.841 |
goodbye | +0.826 |
Neg logit contributions | |
ibaba | -0.534 |
Vector | -0.510 |
racuse | -0.502 |
WARE | -0.485 |
buquerque | -0.481 |
Token | What |
Token ID | 2061 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
arthed | +0.612 |
ept | +0.576 |
##### | +0.563 |
iltr | +0.551 |
Dol | +0.541 |
Neg logit contributions | |
ibaba | -0.504 |
Vector | -0.483 |
racuse | -0.479 |
WARE | -0.465 |
buquerque | -0.463 |
Token | to |
Token ID | 284 |
Feature activation | -4.44e+00 |
Pos logit contributions | |
ibaba | +1.891 |
Vector | +1.836 |
racuse | +1.800 |
WARE | +1.747 |
buquerque | +1.747 |
Neg logit contributions | |
arthed | -2.391 |
ept | -2.249 |
##### | -2.171 |
iltr | -2.141 |
Dol | -2.089 |
Token | head |
Token ID | 1182 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
Vector | +2.821 |
ibaba | +2.800 |
racuse | +2.649 |
WARE | +2.625 |
Startup | +2.604 |
Neg logit contributions | |
arthed | -3.197 |
ept | -2.903 |
iltr | -2.746 |
##### | -2.729 |
Dol | -2.650 |
Token | over |
Token ID | 625 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
ibaba | +1.817 |
Vector | +1.765 |
racuse | +1.713 |
WARE | +1.687 |
buquerque | +1.653 |
Neg logit contributions | |
arthed | -2.674 |
ept | -2.493 |
##### | -2.436 |
iltr | -2.412 |
Dol | -2.342 |
Token | to |
Token ID | 284 |
Feature activation | -6.47e+00 |
Pos logit contributions | |
ibaba | +1.554 |
Vector | +1.504 |
racuse | +1.469 |
WARE | +1.435 |
buquerque | +1.429 |
Neg logit contributions | |
arthed | -1.948 |
ept | -1.814 |
##### | -1.747 |
iltr | -1.736 |
Dol | -1.694 |
Token | http |
Token ID | 2638 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
ibaba | +3.664 |
Vector | +3.641 |
racuse | +3.377 |
WARE | +3.345 |
Startup | +3.330 |
Neg logit contributions | |
arthed | -4.953 |
ept | -4.627 |
iltr | -4.445 |
##### | -4.346 |
Dol | -4.275 |
Token | :// |
Token ID | 1378 |
Feature activation | -7.99e+00 |
Pos logit contributions | |
ibaba | +1.940 |
Vector | +1.897 |
racuse | +1.845 |
WARE | +1.821 |
ABLE | +1.803 |
Neg logit contributions | |
arthed | -2.533 |
ept | -2.348 |
iltr | -2.275 |
##### | -2.260 |
Dol | -2.200 |
Token | store |
Token ID | 8095 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
ibaba | +4.882 |
Vector | +4.544 |
racuse | +4.461 |
WARE | +4.430 |
buquerque | +4.401 |
Neg logit contributions | |
arthed | -5.720 |
ept | -5.219 |
Dol | -5.116 |
##### | -5.108 |
iltr | -5.050 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
arthed | +1.370 |
ept | +1.283 |
##### | +1.257 |
iltr | +1.230 |
Dol | +1.194 |
Neg logit contributions | |
ibaba | -1.225 |
Vector | -1.191 |
racuse | -1.175 |
WARE | -1.152 |
buquerque | -1.137 |
Token | unt |
Token ID | 2797 |
Feature activation | -7.09e+00 |
Pos logit contributions | |
ibaba | +0.069 |
Vector | +0.067 |
racuse | +0.066 |
WARE | +0.065 |
buquerque | +0.064 |
Neg logit contributions | |
arthed | -0.073 |
ept | -0.068 |
##### | -0.066 |
iltr | -0.065 |
Dol | -0.064 |
Token | app |
Token ID | 1324 |
Feature activation | +9.65e-01 |
Pos logit contributions | |
ibaba | +1.821 |
racuse | +1.525 |
buquerque | +1.440 |
Vector | +1.427 |
ABLE | +1.288 |
Neg logit contributions | |
arthed | -7.505 |
##### | -7.153 |
Dol | -7.117 |
ept | -7.113 |
iltr | -7.037 |
Token | d |
Token ID | 67 |
Feature activation | +6.07e+00 |
Pos logit contributions | |
arthed | +0.858 |
ept | +0.811 |
##### | +0.794 |
iltr | +0.782 |
Dol | +0.767 |
Neg logit contributions | |
ibaba | -0.449 |
Vector | -0.430 |
racuse | -0.419 |
WARE | -0.411 |
buquerque | -0.403 |
Token | . |
Token ID | 764 |
Feature activation | -5.78e+00 |
Pos logit contributions | |
Vector | +4.520 |
ibaba | +4.351 |
racuse | +4.180 |
buquerque | +4.124 |
ABLE | +4.112 |
Neg logit contributions | |
arthed | -5.047 |
ept | -4.625 |
iltr | -4.530 |
##### | -4.510 |
Dol | -4.343 |
Token | In |
Token ID | 554 |
Feature activation | -5.47e+00 |
Pos logit contributions | |
Vector | +3.162 |
ibaba | +3.048 |
Higgins | +2.879 |
Startup | +2.864 |
racuse | +2.836 |
Neg logit contributions | |
arthed | -4.629 |
ept | -4.299 |
iltr | -4.201 |
##### | -4.128 |
goodbye | -4.006 |
Token | return |
Token ID | 1441 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
Vector | +3.276 |
ibaba | +3.252 |
WARE | +3.054 |
racuse | +3.037 |
Higgins | +3.031 |
Neg logit contributions | |
arthed | -4.044 |
iltr | -3.643 |
ept | -3.627 |
##### | -3.623 |
Dol | -3.463 |
Token | , |
Token ID | 11 |
Feature activation | -6.23e+00 |
Pos logit contributions | |
ibaba | +1.756 |
Vector | +1.749 |
racuse | +1.671 |
WARE | +1.640 |
buquerque | +1.625 |
Neg logit contributions | |
arthed | -2.463 |
ept | -2.287 |
##### | -2.225 |
iltr | -2.215 |
Dol | -2.142 |
Token | you |
Token ID | 345 |
Feature activation | -9.73e+00 |
Pos logit contributions | |
Vector | +3.503 |
ibaba | +3.435 |
Higgins | +3.222 |
racuse | +3.217 |
WARE | +3.146 |
Neg logit contributions | |
arthed | -4.882 |
ept | -4.438 |
iltr | -4.384 |
##### | -4.371 |
Dol | -4.221 |
Token | get |
Token ID | 651 |
Feature activation | -1.81e+01 |
Pos logit contributions | |
Vector | +5.401 |
ibaba | +5.070 |
Higgins | +4.877 |
racuse | +4.812 |
ABLE | +4.745 |
Neg logit contributions | |
arthed | -7.614 |
ept | -6.749 |
iltr | -6.510 |
Dol | -6.371 |
##### | -6.333 |
Token | an |
Token ID | 281 |
Feature activation | -1.29e+01 |
Pos logit contributions | |
Vector | +9.295 |
Startup | +8.036 |
Hamilton | +8.035 |
unsafe | +7.934 |
Higgins | +7.910 |
Neg logit contributions | |
arthed | -13.302 |
iltr | -11.902 |
ept | -11.654 |
##### | -11.212 |
brate | -11.146 |
Token | optional |
Token ID | 11902 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
Vector | +7.665 |
unsafe | +6.981 |
sidx | +6.953 |
ibaba | +6.879 |
Startup | +6.826 |
Neg logit contributions | |
arthed | -9.028 |
ept | -7.994 |
iltr | -7.978 |
brate | -7.856 |
##### | -7.802 |
Token | Shared |
Token ID | 39403 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
Vector | +6.144 |
Startup | +5.531 |
Hamilton | +5.448 |
Higgins | +5.429 |
ibaba | +5.413 |
Neg logit contributions | |
arthed | -7.505 |
iltr | -6.690 |
ept | -6.652 |
##### | -6.398 |
Dol | -6.187 |
Token | pointer |
Token ID | 17562 |
Feature activation | -5.65e+00 |
Pos logit contributions | |
ibaba | +1.332 |
Vector | +1.320 |
racuse | +1.280 |
WARE | +1.255 |
buquerque | +1.244 |
Neg logit contributions | |
arthed | -1.456 |
ept | -1.340 |
##### | -1.318 |
iltr | -1.302 |
Dol | -1.263 |
Token | back |
Token ID | 736 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Vector | +3.206 |
ibaba | +3.056 |
racuse | +2.951 |
Higgins | +2.928 |
ABLE | +2.928 |
Neg logit contributions | |
arthed | -4.401 |
ept | -4.043 |
iltr | -3.984 |
##### | -3.954 |
Dol | -3.801 |
Token | your |
Token ID | 534 |
Feature activation | -1.15e+01 |
Pos logit contributions | |
Vector | +4.496 |
ibaba | +4.305 |
racuse | +4.044 |
Higgins | +4.039 |
buquerque | +4.038 |
Neg logit contributions | |
arthed | -6.814 |
ept | -6.379 |
iltr | -6.134 |
##### | -6.043 |
ⓘ | -5.843 |
Token | daughter |
Token ID | 4957 |
Feature activation | +7.06e-01 |
Pos logit contributions | |
Vector | +6.195 |
Hamilton | +5.674 |
Higgins | +5.614 |
Startup | +5.584 |
viruses | +5.504 |
Neg logit contributions | |
arthed | -8.895 |
ept | -8.169 |
iltr | -7.794 |
##### | -7.528 |
ⓘ | -7.439 |
Token | take |
Token ID | 1011 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
arthed | +0.503 |
ept | +0.472 |
##### | +0.463 |
iltr | +0.453 |
Dol | +0.445 |
Neg logit contributions | |
ibaba | -0.450 |
Vector | -0.431 |
racuse | -0.427 |
WARE | -0.419 |
buquerque | -0.416 |
Token | instruction |
Token ID | 12064 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
Vector | +6.590 |
Hamilton | +6.043 |
ibaba | +5.960 |
Higgins | +5.935 |
Startup | +5.882 |
Neg logit contributions | |
arthed | -10.334 |
ept | -9.724 |
iltr | -9.243 |
##### | -9.063 |
igne | -8.729 |
Token | from |
Token ID | 422 |
Feature activation | -1.21e+01 |
Pos logit contributions | |
Vector | +5.359 |
ibaba | +5.108 |
WARE | +4.829 |
Higgins | +4.782 |
Startup | +4.769 |
Neg logit contributions | |
arthed | -9.409 |
ept | -8.661 |
iltr | -8.166 |
##### | -8.159 |
ⓘ | -8.019 |
Token | these |
Token ID | 777 |
Feature activation | -1.70e+01 |
Pos logit contributions | |
Vector | +6.339 |
Hamilton | +5.737 |
Higgins | +5.724 |
ibaba | +5.631 |
Notting | +5.626 |
Neg logit contributions | |
arthed | -9.386 |
ept | -8.626 |
iltr | -8.558 |
##### | -8.166 |
ⓘ | -8.155 |
Token | videos |
Token ID | 5861 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
Vector | +8.393 |
Hamilton | +7.859 |
Higgins | +7.843 |
Startup | +7.659 |
viruses | +7.659 |
Neg logit contributions | |
arthed | -12.449 |
iltr | -11.227 |
ept | -11.218 |
ⓘ | -10.812 |
brate | -10.601 |
Token | ? |
Token ID | 30 |
Feature activation | -9.38e+00 |
Pos logit contributions | |
ibaba | +1.462 |
Vector | +1.432 |
racuse | +1.361 |
WARE | +1.354 |
buquerque | +1.324 |
Neg logit contributions | |
arthed | -2.732 |
ept | -2.571 |
##### | -2.503 |
iltr | -2.487 |
Dol | -2.407 |
Token |
|
Token ID | 198 |
Feature activation | -8.24e+00 |
Pos logit contributions | |
Vector | +4.704 |
Higgins | +4.406 |
Startup | +4.334 |
Hamilton | +4.216 |
ibaba | +4.130 |
Neg logit contributions | |
arthed | -7.922 |
ept | -7.377 |
iltr | -7.078 |
onement | -6.726 |
goodbye | -6.714 |
Token |
|
Token ID | 198 |
Feature activation | -7.44e+00 |
Pos logit contributions | |
Vector | +3.585 |
ABLE | +3.320 |
Hamilton | +3.259 |
Startup | +3.250 |
Higgins | +3.250 |
Neg logit contributions | |
arthed | -7.324 |
ept | -6.829 |
iltr | -6.761 |
agements | -6.362 |
##### | -6.350 |
Token | Image |
Token ID | 5159 |
Feature activation | -8.30e+00 |
Pos logit contributions | |
Vector | +4.026 |
WARE | +3.994 |
ABLE | +3.962 |
ibaba | +3.927 |
Hamilton | +3.831 |
Neg logit contributions | |
arthed | -5.623 |
ept | -5.232 |
iltr | -5.089 |
##### | -5.078 |
goodbye | -4.944 |
Token | sentence |
Token ID | 6827 |
Feature activation | -6.56e+00 |
Pos logit contributions | |
Vector | +7.253 |
ibaba | +7.198 |
Higgins | +7.069 |
ABLE | +6.908 |
Hamilton | +6.871 |
Neg logit contributions | |
arthed | -6.299 |
ept | -5.528 |
iltr | -5.368 |
##### | -5.119 |
agements | -4.888 |
Token | . |
Token ID | 13 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
Vector | +3.466 |
ibaba | +3.446 |
Higgins | +3.310 |
racuse | +3.261 |
buquerque | +3.235 |
Neg logit contributions | |
arthed | -5.305 |
ept | -4.858 |
iltr | -4.751 |
##### | -4.709 |
Dol | -4.469 |
Token | Many |
Token ID | 4650 |
Feature activation | +4.86e+00 |
Pos logit contributions | |
ibaba | +1.533 |
Vector | +1.522 |
racuse | +1.449 |
Higgins | +1.445 |
WARE | +1.439 |
Neg logit contributions | |
arthed | -2.078 |
ept | -1.949 |
iltr | -1.882 |
##### | -1.870 |
goodbye | -1.826 |
Token | found |
Token ID | 1043 |
Feature activation | -5.35e+00 |
Pos logit contributions | |
arthed | +3.314 |
##### | +3.135 |
ept | +3.132 |
Dol | +3.031 |
iltr | +2.953 |
Neg logit contributions | |
ibaba | -3.221 |
racuse | -3.005 |
WARE | -2.962 |
Vector | -2.943 |
buquerque | -2.901 |
Token | it |
Token ID | 340 |
Feature activation | -1.23e+01 |
Pos logit contributions | |
ibaba | +3.062 |
Vector | +3.056 |
racuse | +2.956 |
Higgins | +2.931 |
WARE | +2.852 |
Neg logit contributions | |
arthed | -3.984 |
ept | -3.739 |
##### | -3.643 |
iltr | -3.640 |
goodbye | -3.452 |
Token | quite |
Token ID | 2407 |
Feature activation | -1.66e+01 |
Pos logit contributions | |
Vector | +5.694 |
Higgins | +5.656 |
ibaba | +5.614 |
unsafe | +5.565 |
Hamilton | +5.493 |
Neg logit contributions | |
arthed | -9.887 |
ept | -9.307 |
iltr | -9.142 |
##### | -8.840 |
brate | -8.415 |
Token | suspicious |
Token ID | 13678 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
Higgins | +7.125 |
unsafe | +7.106 |
Vector | +6.873 |
Hamilton | +6.791 |
viruses | +6.707 |
Neg logit contributions | |
arthed | -12.790 |
ept | -12.274 |
iltr | -12.184 |
##### | -11.644 |
olon | -11.590 |
Token | that |
Token ID | 326 |
Feature activation | +6.41e-01 |
Pos logit contributions | |
ibaba | +1.659 |
Vector | +1.614 |
racuse | +1.548 |
WARE | +1.539 |
Higgins | +1.521 |
Neg logit contributions | |
arthed | -2.130 |
ept | -1.984 |
##### | -1.937 |
iltr | -1.918 |
goodbye | -1.847 |
Token | the |
Token ID | 262 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
arthed | +0.438 |
ept | +0.409 |
##### | +0.400 |
iltr | +0.391 |
Dol | +0.383 |
Neg logit contributions | |
ibaba | -0.429 |
Vector | -0.412 |
racuse | -0.408 |
WARE | -0.400 |
buquerque | -0.397 |
Token | officer |
Token ID | 3818 |
Feature activation | +5.67e+00 |
Pos logit contributions | |
arthed | +0.133 |
ept | +0.124 |
##### | +0.122 |
iltr | +0.119 |
Dol | +0.117 |
Neg logit contributions | |
ibaba | -0.109 |
Vector | -0.104 |
racuse | -0.103 |
WARE | -0.101 |
buquerque | -0.100 |
Token | who |
Token ID | 508 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
arthed | +3.733 |
ept | +3.543 |
##### | +3.487 |
Dol | +3.440 |
goodbye | +3.337 |
Neg logit contributions | |
ibaba | -3.867 |
racuse | -3.665 |
Vector | -3.549 |
buquerque | -3.528 |
WARE | -3.520 |
Token |
|
Token ID | 198 |
Feature activation | +4.22e-01 |
Pos logit contributions | |
ibaba | +2.275 |
Vector | +2.216 |
racuse | +2.155 |
WARE | +2.111 |
buquerque | +2.097 |
Neg logit contributions | |
arthed | -3.197 |
ept | -3.018 |
iltr | -2.915 |
##### | -2.913 |
goodbye | -2.814 |
Token |
|
Token ID | 198 |
Feature activation | -5.74e+00 |
Pos logit contributions | |
arthed | +0.308 |
ept | +0.289 |
##### | +0.283 |
iltr | +0.276 |
Dol | +0.271 |
Neg logit contributions | |
ibaba | -0.263 |
Vector | -0.252 |
racuse | -0.249 |
WARE | -0.244 |
buquerque | -0.243 |
Token | 7 |
Token ID | 22 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
ibaba | +3.021 |
Vector | +2.909 |
ABLE | +2.873 |
WARE | +2.873 |
racuse | +2.862 |
Neg logit contributions | |
arthed | -4.474 |
ept | -4.255 |
iltr | -4.159 |
##### | -4.158 |
Dol | -4.050 |
Token | . |
Token ID | 13 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
ibaba | +0.243 |
Vector | +0.234 |
racuse | +0.230 |
WARE | +0.225 |
buquerque | +0.223 |
Neg logit contributions | |
arthed | -0.294 |
ept | -0.275 |
##### | -0.268 |
iltr | -0.264 |
Dol | -0.257 |
Token | Next |
Token ID | 7406 |
Feature activation | +4.53e+00 |
Pos logit contributions | |
ibaba | +0.662 |
Vector | +0.645 |
racuse | +0.628 |
WARE | +0.616 |
buquerque | +0.612 |
Neg logit contributions | |
arthed | -0.808 |
ept | -0.756 |
##### | -0.736 |
iltr | -0.727 |
goodbye | -0.703 |
Token | of |
Token ID | 286 |
Feature activation | -1.69e+01 |
Pos logit contributions | |
arthed | +3.242 |
ept | +3.079 |
##### | +3.019 |
Dol | +2.898 |
iltr | +2.894 |
Neg logit contributions | |
ibaba | -2.819 |
racuse | -2.692 |
WARE | -2.683 |
Vector | -2.614 |
buquerque | -2.606 |
Token | Kin |
Token ID | 16645 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
ibaba | +8.389 |
Vector | +8.072 |
Alone | +7.816 |
Startup | +7.793 |
Hamilton | +7.688 |
Neg logit contributions | |
arthed | -12.454 |
iltr | -11.334 |
goodbye | -11.094 |
ept | -10.957 |
CTV | -10.806 |
Token | 03 |
Token ID | 7643 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
ibaba | +1.830 |
Vector | +1.783 |
racuse | +1.741 |
buquerque | +1.710 |
WARE | +1.700 |
Neg logit contributions | |
arthed | -1.484 |
ept | -1.353 |
##### | -1.317 |
iltr | -1.306 |
goodbye | -1.256 |
Token | : |
Token ID | 25 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
ibaba | +0.210 |
Vector | +0.202 |
racuse | +0.199 |
WARE | +0.195 |
buquerque | +0.193 |
Neg logit contributions | |
arthed | -0.252 |
ept | -0.236 |
##### | -0.230 |
iltr | -0.227 |
Dol | -0.221 |
Token | 06 |
Token ID | 3312 |
Feature activation | -4.87e+00 |
Pos logit contributions | |
arthed | +1.213 |
ept | +1.135 |
##### | +1.122 |
iltr | +1.078 |
Dol | +1.067 |
Neg logit contributions | |
ibaba | -1.293 |
racuse | -1.221 |
Vector | -1.219 |
buquerque | -1.201 |
WARE | -1.192 |
Token |
|
Token ID | 198 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
ibaba | +2.641 |
Vector | +2.580 |
racuse | +2.495 |
WARE | +2.448 |
buquerque | +2.429 |
Neg logit contributions | |
arthed | -3.830 |
ept | -3.619 |
iltr | -3.500 |
##### | -3.480 |
goodbye | -3.380 |
Token | The |
Token ID | 464 |
Feature activation | -5.54e+00 |
Pos logit contributions | |
arthed | +0.620 |
ept | +0.581 |
##### | +0.568 |
iltr | +0.556 |
Dol | +0.544 |
Neg logit contributions | |
ibaba | -0.530 |
Vector | -0.508 |
racuse | -0.502 |
WARE | -0.488 |
buquerque | -0.486 |
Token | bill |
Token ID | 2855 |
Feature activation | -6.49e+00 |
Pos logit contributions | |
ibaba | +3.291 |
Vector | +3.288 |
racuse | +3.128 |
Higgins | +3.105 |
Startup | +3.096 |
Neg logit contributions | |
arthed | -4.065 |
ept | -3.753 |
iltr | -3.644 |
##### | -3.622 |
ⓘ | -3.453 |
Token | will |
Token ID | 481 |
Feature activation | -5.02e+00 |
Pos logit contributions | |
ibaba | +3.490 |
Vector | +3.457 |
Higgins | +3.334 |
buquerque | +3.229 |
racuse | +3.217 |
Neg logit contributions | |
arthed | -5.165 |
ept | -4.752 |
iltr | -4.626 |
##### | -4.558 |
CTV | -4.369 |
Token | face |
Token ID | 1986 |
Feature activation | -5.47e+00 |
Pos logit contributions | |
ibaba | +3.251 |
Vector | +3.246 |
Higgins | +3.051 |
racuse | +3.047 |
Startup | +3.008 |
Neg logit contributions | |
arthed | -3.553 |
ept | -3.240 |
iltr | -3.098 |
##### | -3.044 |
Dol | -2.929 |
Token | an |
Token ID | 281 |
Feature activation | -4.53e+00 |
Pos logit contributions | |
ibaba | +3.180 |
Vector | +3.088 |
Higgins | +2.980 |
racuse | +2.939 |
buquerque | +2.883 |
Neg logit contributions | |
arthed | -4.114 |
ept | -3.890 |
iltr | -3.761 |
##### | -3.741 |
ⓘ | -3.532 |
Token | uphill |
Token ID | 35256 |
Feature activation | -1.75e+01 |
Pos logit contributions | |
ibaba | +2.389 |
Vector | +2.276 |
Higgins | +2.178 |
racuse | +2.155 |
buquerque | +2.133 |
Neg logit contributions | |
arthed | -3.722 |
ept | -3.456 |
##### | -3.371 |
iltr | -3.328 |
Dol | -3.227 |
Token | battle |
Token ID | 3344 |
Feature activation | -9.20e+00 |
Pos logit contributions | |
ibaba | +6.556 |
Vector | +6.525 |
Hamilton | +6.116 |
Higgins | +5.829 |
Notting | +5.692 |
Neg logit contributions | |
arthed | -15.192 |
ept | -13.746 |
iltr | -13.644 |
auna | -13.292 |
brate | -13.124 |
Token | in |
Token ID | 287 |
Feature activation | +5.52e-03 |
Pos logit contributions | |
ibaba | +4.623 |
Vector | +4.574 |
Higgins | +4.290 |
WARE | +4.247 |
Hamilton | +4.240 |
Neg logit contributions | |
arthed | -7.362 |
ept | -6.965 |
iltr | -6.621 |
##### | -6.583 |
brate | -6.359 |
Token | the |
Token ID | 262 |
Feature activation | +3.59e+00 |
Pos logit contributions | |
arthed | +0.004 |
ept | +0.004 |
##### | +0.004 |
iltr | +0.004 |
Dol | +0.004 |
Neg logit contributions | |
ibaba | -0.003 |
Vector | -0.003 |
racuse | -0.003 |
WARE | -0.003 |
buquerque | -0.003 |
Token | Senate |
Token ID | 3845 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
arthed | +2.594 |
ept | +2.462 |
##### | +2.436 |
Dol | +2.371 |
iltr | +2.350 |
Neg logit contributions | |
ibaba | -2.235 |
racuse | -2.149 |
Vector | -2.072 |
WARE | -2.064 |
buquerque | -2.031 |
Token | . |
Token ID | 13 |
Feature activation | +3.79e+00 |
Pos logit contributions | |
ibaba | +0.958 |
Vector | +0.925 |
racuse | +0.897 |
WARE | +0.886 |
buquerque | +0.878 |
Neg logit contributions | |
arthed | -1.222 |
ept | -1.143 |
##### | -1.112 |
iltr | -1.095 |
Dol | -1.065 |